{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Compare PLV3 log mode results with dataset used to do calibration. Use the Jun-Sep2019 dataset.\n",
"Using the Population Stability Index (PSI), evaluate:\n",
"- Feature distributions\n",
"- Score distribution"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import lightgbm\n",
"import pymysql\n",
"import pickle\n",
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import configparser\n",
"config = configparser.ConfigParser()\n",
"config.read('/home/ec2-user/SageMaker/zhilal/config.ini')\n",
"\n",
"host = config['MYSQL-ROOT']['HOST']\n",
"user = config['MYSQL-ROOT']['USER']\n",
"password = config['MYSQL-ROOT']['PASSWORD']\n",
"\n",
"def connect_sql():\n",
" cnx = pymysql.connect(host=host,\n",
" user=user,\n",
" password=password,\n",
" cursorclass=pymysql.cursors.DictCursor)\n",
" return cnx"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"log_mode_q = '''\n",
"SELECT *\n",
"FROM ds.b_score_pl_log\n",
"WHERE calculation_date > '2020-02-21'\n",
"AND engine_name = 'personal loan v3'\n",
"'''"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"log_mode_df = pd.read_sql(log_mode_q, connect_sql())"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(30758, 8)"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"log_mode_df.shape"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# remove duplicates, keep the last score calculated\n",
"log_mode_df = log_mode_df.sort_values(by=['trx_id', 'user_id', 'calculation_date'], ascending=True)\\\n",
" .drop_duplicates(subset=['trx_id', 'user_id'], keep='last').reset_index(drop=True)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(30502, 8)"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"log_mode_df.shape"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"30502"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"log_mode_df['trx_id'].nunique()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"nan = np.nan\n",
"log_mode_df['feature'] = log_mode_df['feature'].apply(lambda z: eval(z))"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"log_mode_feats = pd.DataFrame(log_mode_df['feature'].tolist())\n",
"log_mode_df = log_mode_df.merge(log_mode_feats, how='left', left_index=True, right_index=True)"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"log_mode_df"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Load June - September 2019 Data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Load Original Dataset"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"base_df = pd.read_parquet('data/interim/pl_jumbo_mini_junjulaugsep_27012020.parquet')\\\n",
" .rename(columns={'loan_amount':'amount'})\n",
"pl_feats = pd.read_parquet('data/interim/pl_model_v3_feats_27012020.parquet')\n",
"\n",
"# REMOVE OLD util_non_pl\n",
"pl_feats = pl_feats.drop(columns=['util_non_pl'])\n",
"# ATTACH NEW util_non_pl\n",
"fixed_util_non_pl = pd.read_parquet('data/interim/fixed_util_non_pl.parquet')\n",
"pl_feats = pl_feats.merge(fixed_util_non_pl, how='left', on=['user_id', 'trx_id'])\n",
"\n",
"pf_feats = pd.read_parquet('data/interim/pl_model_v3_pefindo_feats_27012020.parquet')\\\n",
" .drop(columns=['transaction_date'])\n",
"new_feats = pd.read_parquet('data/interim/pl_model_v3_new_feats_27012020.parquet')\n",
"dgtl_trx_feats = pd.read_parquet('data/interim/pl_model_v3_dgtl_trx_feats_18022020.parquet')\n",
"new_trx_feats = pd.read_parquet('data/interim/pl_model_v3_new_trx_feats_27012020.parquet')"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(150672, 18)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(150672, 28)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(150672, 9)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(150672, 85)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(150672, 56)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(150672, 30)"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"display(base_df.shape)\n",
"display(pl_feats.shape)\n",
"display(pf_feats.shape)\n",
"display(new_feats.shape)\n",
"display(dgtl_trx_feats.shape)\n",
"display(new_trx_feats.shape)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"raw_df = base_df.merge(pl_feats, how='left', on=['user_id', 'trx_id'])\\\n",
" .merge(pf_feats, how='left', on=['user_id', 'trx_id'])\\\n",
" .merge(new_feats, how='left', on=['user_id', 'trx_id'])\\\n",
" .merge(dgtl_trx_feats, how='left', on=['user_id', 'trx_id'])\\\n",
" .merge(new_trx_feats, how='left', on=['user_id', 'trx_id'])"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"rename_feats = {\n",
" 'denied_trx_co_3mo':'trx_denied_co_90',\n",
" 'delin_max_dpd_3mo':'delin_max_dpd_90',\n",
" 'pl_trx_co_jumbomini_4_90':'pl_trx_suc_co_90',\n",
" 'settle_to_due_last_pl_180':'pl_settle_to_due_last_180',\n",
" 'settled_trx_sum_1mo':'trx_sett_sum_30',\n",
" 'td_date':'date_of_month',\n",
" 'time_approve_to_pl_hour':'time_appr_to_pl_hour',\n",
" 'time_from_prev_pl_settle':'time_from_last_sett_pl_hour'\n",
"}\n",
"\n",
"raw_df = raw_df.rename(columns=rename_feats)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Load Rejected Users"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"rej_base = pd.read_parquet('data/pl_rejected_users_junjulaugsep_04022020.parquet')\\\n",
" .rename(columns={'transaction_id':'trx_id'})\n",
"rej_feats = pd.read_parquet('data/pl_rejected_users_feats_04022020.parquet')\n",
"# REMOVE util_non_pl & nondgtl_nonpl_trx_co_180\n",
"rej_feats = rej_feats.drop(columns=['util_non_pl', 'nondgtl_nonpl_trx_co_180'])\n",
"# ATTACH fixed feats\n",
"fixed_rej_util = pd.read_parquet('data/fixed_pl_rejected_util_non_pl.parquet')\n",
"fixed_rej_nondgtl = pd.read_parquet('data/fixed_pl_rejected_dgtl_trx_feats.parquet')\n",
"rej_feats = rej_feats.merge(fixed_rej_util[['user_id', 'trx_id', 'util_non_pl']], how='left', \n",
" on=['user_id', 'trx_id'])\\\n",
" .merge(fixed_rej_nondgtl[['user_id', 'trx_id', 'nondgtl_nonpl_trx_co_180']], how='left', \n",
" on=['user_id', 'trx_id'])\n",
"\n",
"rej_pef_feats = pd.read_parquet('data/pl_rejected_users_pefindo_feats_04022020.parquet')\\\n",
" .rename(columns={'transaction_id':'trx_id'})"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
"cols_to_drop = [\n",
" 'rule_name', 'result', 'remarks',\n",
" 'transaction_date', 'amount', 'performance_window', 'max_dpd'\n",
"]"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"rej_feats = rej_feats.drop(columns=cols_to_drop)\n",
"rej_pef_feats = rej_pef_feats.drop(columns=['transaction_date'])"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(16190, 10)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(16190, 191)"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(16190, 9)"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"display(rej_base.shape)\n",
"display(rej_feats.shape)\n",
"display(rej_pef_feats.shape)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"rej_raw = rej_base.merge(rej_feats, how='left', on=['user_id', 'trx_id'])\\\n",
" .merge(rej_pef_feats, how='left', on=['user_id', 'trx_id'])"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"rename_feats = {\n",
" 'denied_trx_co_3mo':'trx_denied_co_90',\n",
" 'delin_max_dpd_3mo':'delin_max_dpd_90',\n",
" 'pl_trx_co_jumbomini_4_90':'pl_trx_suc_co_90',\n",
" 'settle_to_due_last_pl_180':'pl_settle_to_due_last_180',\n",
" 'settled_trx_sum_1mo':'trx_sett_sum_30',\n",
" 'td_date':'date_of_month',\n",
" 'time_approve_to_pl_hour':'time_appr_to_pl_hour',\n",
" 'time_from_prev_pl_settle':'time_from_last_sett_pl_hour'\n",
"}\n",
"\n",
"rej_raw = rej_raw.rename(columns=rename_feats)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Concat Original Data with Rejected Users Data"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"features = ['ap_co_plo',\n",
" 'current_dpd',\n",
" 'date_of_month',\n",
" 'delin_max_dpd_90',\n",
" 'nondgtl_nonpl_trx_co_180',\n",
" 'oth_last_rep_days',\n",
" 'oth_last_rep_dpd',\n",
" 'payment_type',\n",
" 'pf_delin_max_dpd_12mo',\n",
" 'pl_settle_to_due_last_180',\n",
" 'pl_trx_suc_co_90',\n",
" 'time_appr_to_pl_hour',\n",
" 'time_from_last_sett_pl_hour',\n",
" 'trx_denied_co_90',\n",
" 'trx_sett_sum_30',\n",
" 'util_non_pl',\n",
" 'util_pl'\n",
" ]\n",
"\n",
"features = sorted(features)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [],
"source": [
"calib_df = pd.concat([raw_df[features], rej_raw[features]])\\\n",
" .reset_index(drop=True)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(166862, 17)"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"calib_df.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Feature Transformations"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"calib_df['payment_type'] = np.where(calib_df['payment_type'] == '3_months', 0, 1)"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"# NaN values in time_approve_to_pl_hour are caused by missing approval time in DB\n",
"# fill with 0. (There are only 4 cases)\n",
"calib_df['time_appr_to_pl_hour'] = calib_df['time_appr_to_pl_hour'].fillna(2348.0)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"# Negative values caused by error in loan table. User's previous pl\n",
"# settlement date is after the current pl transaction date.\n",
"# Replace with median values. (There are only 2 cases)\n",
"neg_index = calib_df[calib_df['time_from_last_sett_pl_hour'] < 0].index\n",
"med_val = 17.0\n",
"calib_df.loc[neg_index, 'time_from_last_sett_pl_hour'] = med_val"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Calculate Score"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [],
"source": [
"mod_lgb = pickle.load(open('lgbm_pl_score_v3.p', 'rb'))"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"1079"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mod_lgb.best_iteration_"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [],
"source": [
"calib_df['score'] = mod_lgb.predict_proba(calib_df[features], \n",
" num_iteration=mod_lgb.best_iteration_\n",
" )[:,1]"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [],
"source": [
"calib_df['score'] = calib_df['score'].apply(lambda x: 0.0045+0.622*x)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Compare Training Set with Log Mode Data"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [],
"source": [
"def create_psi_bins(df, feature, bins):\n",
" _, bin_array = pd.qcut(df[feature], 10, duplicates='drop', retbins=True)\n",
" bin_array[0] = np.NINF\n",
" bin_array[-1] = np.inf\n",
" return bin_array.tolist()"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"discrete_cols = ['date_of_month',\n",
" 'payment_type',\n",
" ]\n",
"relevant_feats = sorted(list(set(features) - set(discrete_cols))) + ['score']"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [],
"source": [
"feature_bins_psi = {}\n",
"for feat in relevant_feats:\n",
" feature_bins_psi[feat] = create_psi_bins(calib_df, feat, 10)\n",
"for feat in discrete_cols:\n",
" feature_bins_psi[feat] = []"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
"{'ap_co_plo': [-inf, 1.0, 2.0, inf],\n",
" 'current_dpd': [-inf,\n",
" -30.0,\n",
" -27.0,\n",
" -24.0,\n",
" -19.0,\n",
" -15.0,\n",
" -10.0,\n",
" -6.0,\n",
" -3.0,\n",
" 0.0,\n",
" inf],\n",
" 'delin_max_dpd_90': [-inf, -14.0, -4.0, -2.0, -1.0, 0.0, 1.0, 3.0, 9.0, inf],\n",
" 'nondgtl_nonpl_trx_co_180': [-inf,\n",
" 1.0,\n",
" 2.0,\n",
" 3.0,\n",
" 6.0,\n",
" 10.0,\n",
" 18.0,\n",
" 34.0,\n",
" 84.0,\n",
" inf],\n",
" 'oth_last_rep_days': [-inf, 1.0, 2.0, 4.0, 8.0, 16.0, inf],\n",
" 'oth_last_rep_dpd': [-inf,\n",
" -35.0,\n",
" -26.0,\n",
" -19.0,\n",
" -12.0,\n",
" -7.0,\n",
" -4.0,\n",
" -2.0,\n",
" 0.0,\n",
" 1.0,\n",
" inf],\n",
" 'pf_delin_max_dpd_12mo': [-inf, 0.0, 3.0, 19.0, 67.0, inf],\n",
" 'pl_settle_to_due_last_180': [-inf,\n",
" 1.0,\n",
" 5.0,\n",
" 12.0,\n",
" 20.0,\n",
" 28.0,\n",
" 40.0,\n",
" 60.0,\n",
" 83.0,\n",
" 118.0,\n",
" inf],\n",
" 'pl_trx_suc_co_90': [-inf, 1.0, inf],\n",
" 'time_appr_to_pl_hour': [-inf,\n",
" 3.0,\n",
" 65.0,\n",
" 311.0,\n",
" 852.0,\n",
" 2154.0,\n",
" 3717.0,\n",
" 5238.700000000012,\n",
" 7565.0,\n",
" 11549.0,\n",
" inf],\n",
" 'time_from_last_sett_pl_hour': [-inf,\n",
" 0.04416666666666667,\n",
" 0.10888888888888888,\n",
" 0.39361111111111113,\n",
" 3.6312222222222226,\n",
" 20.22111111111111,\n",
" 65.87494444444452,\n",
" 187.34288888888895,\n",
" 526.391388888889,\n",
" 1446.3358888888918,\n",
" inf],\n",
" 'trx_denied_co_90': [-inf, 1.0, 2.0, 3.0, 6.0, inf],\n",
" 'trx_sett_sum_30': [-inf,\n",
" 14580.000000000291,\n",
" 81058.5,\n",
" 172486.4000000002,\n",
" 328431.80000000005,\n",
" 550000.0,\n",
" 1000533.3000000003,\n",
" inf],\n",
" 'util_non_pl': [-inf, 0.13235260000000013, 0.2639, 0.415, 0.592961, inf],\n",
" 'util_pl': [-inf,\n",
" 0.212766,\n",
" 0.30303,\n",
" 0.384615,\n",
" 0.47619,\n",
" 0.57971,\n",
" 0.714286,\n",
" 0.862069,\n",
" 0.945946,\n",
" 0.952381,\n",
" inf],\n",
" 'score': [-inf,\n",
" 0.008545702048863216,\n",
" 0.011138649888654728,\n",
" 0.014542845890260708,\n",
" 0.019319271923095082,\n",
" 0.025971134003191836,\n",
" 0.03548055630670929,\n",
" 0.04956936090114326,\n",
" 0.07338120518788314,\n",
" 0.12258553688502469,\n",
" inf],\n",
" 'date_of_month': [],\n",
" 'payment_type': []}"
]
},
"execution_count": 36,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"feature_bins_psi"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [],
"source": [
"def bin_features(df, bin_dict):\n",
" temp_ls = []\n",
" for k,v in bin_dict.items():\n",
" if len(v) > 0:\n",
" binned = pd.cut(df[k], v)\n",
" temp_ls.append(binned)\n",
" elif len(v) == 0:\n",
" temp_ls.append(df[k])\n",
" output = pd.concat(temp_ls, axis=1)\n",
" return output"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [],
"source": [
"base_psi_binned = bin_features(calib_df, feature_bins_psi)"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(166862, 18)"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"calib_df.shape"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(166862, 18)"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"base_psi_binned.shape"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"prod_psi_binned = bin_features(log_mode_df, feature_bins_psi)"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(30502, 30)"
]
},
"execution_count": 42,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"log_mode_df.shape"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(30502, 18)"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"prod_psi_binned.shape"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [],
"source": [
"def aggregate_bins(df, feature):\n",
" base_psi = df[feature].value_counts(dropna=False).to_frame().sort_index()\n",
" base_psi['perc'] = base_psi[feature]/df.shape[0]\n",
" base_psi.columns = ['count', feature]\n",
" base_psi = base_psi[[feature]]\n",
" return base_psi"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [],
"source": [
"def calculate_psi(base, prod, feature):\n",
" base_psi = aggregate_bins(base, feature)\n",
" prod_psi = aggregate_bins(prod, feature)\n",
" \n",
" psi = base_psi.merge(prod_psi, how='left', left_index=True, right_index=True)\n",
" psi.columns = [feature+'_base', feature+'_prod']\n",
" \n",
" psi['base-prod'] = psi[feature+'_base'] - psi[feature+'_prod']\n",
" psi['ln(base/prod)'] = np.where(np.isinf(np.log(psi[feature+'_base']/psi[feature+'_prod'])), \n",
" 10, np.log(psi[feature+'_base']/psi[feature+'_prod']))\n",
" psi['index'] = psi['base-prod']*psi['ln(base/prod)']\n",
" print(f'{feature} PSI:', psi['index'].sum())\n",
" display(psi)\n",
" psi[[feature+'_base', feature+'_prod']].plot(kind='bar')\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"ap_co_plo PSI: 0.031731790924998074\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>ap_co_plo_base</th>\n",
" <th>ap_co_plo_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.856211</td>\n",
" <td>0.788473</td>\n",
" <td>0.067738</td>\n",
" <td>0.082418</td>\n",
" <td>0.005583</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 2.0]</th>\n",
" <td>0.055908</td>\n",
" <td>0.081896</td>\n",
" <td>-0.025988</td>\n",
" <td>-0.381737</td>\n",
" <td>0.009921</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(2.0, inf]</th>\n",
" <td>0.087881</td>\n",
" <td>0.129631</td>\n",
" <td>-0.041750</td>\n",
" <td>-0.388707</td>\n",
" <td>0.016228</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" ap_co_plo_base ap_co_plo_prod base-prod ln(base/prod) \\\n",
"(-inf, 1.0] 0.856211 0.788473 0.067738 0.082418 \n",
"(1.0, 2.0] 0.055908 0.081896 -0.025988 -0.381737 \n",
"(2.0, inf] 0.087881 0.129631 -0.041750 -0.388707 \n",
"\n",
" index \n",
"(-inf, 1.0] 0.005583 \n",
"(1.0, 2.0] 0.009921 \n",
"(2.0, inf] 0.016228 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"current_dpd PSI: 0.11119985808908078\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>current_dpd_base</th>\n",
" <th>current_dpd_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, -30.0]</th>\n",
" <td>0.074499</td>\n",
" <td>0.037604</td>\n",
" <td>0.036895</td>\n",
" <td>0.683669</td>\n",
" <td>0.025224</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-30.0, -27.0]</th>\n",
" <td>0.081936</td>\n",
" <td>0.086814</td>\n",
" <td>-0.004878</td>\n",
" <td>-0.057830</td>\n",
" <td>0.000282</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-27.0, -24.0]</th>\n",
" <td>0.055914</td>\n",
" <td>0.042423</td>\n",
" <td>0.013491</td>\n",
" <td>0.276122</td>\n",
" <td>0.003725</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-24.0, -19.0]</th>\n",
" <td>0.080815</td>\n",
" <td>0.057078</td>\n",
" <td>0.023737</td>\n",
" <td>0.347743</td>\n",
" <td>0.008254</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-19.0, -15.0]</th>\n",
" <td>0.062045</td>\n",
" <td>0.048030</td>\n",
" <td>0.014016</td>\n",
" <td>0.256046</td>\n",
" <td>0.003589</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-15.0, -10.0]</th>\n",
" <td>0.077819</td>\n",
" <td>0.063307</td>\n",
" <td>0.014511</td>\n",
" <td>0.206382</td>\n",
" <td>0.002995</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-10.0, -6.0]</th>\n",
" <td>0.068961</td>\n",
" <td>0.069733</td>\n",
" <td>-0.000772</td>\n",
" <td>-0.011132</td>\n",
" <td>0.000009</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-6.0, -3.0]</th>\n",
" <td>0.064323</td>\n",
" <td>0.061275</td>\n",
" <td>0.003048</td>\n",
" <td>0.048545</td>\n",
" <td>0.000148</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-3.0, 0.0]</th>\n",
" <td>0.090326</td>\n",
" <td>0.091732</td>\n",
" <td>-0.001406</td>\n",
" <td>-0.015441</td>\n",
" <td>0.000022</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0, inf]</th>\n",
" <td>0.047602</td>\n",
" <td>0.118353</td>\n",
" <td>-0.070751</td>\n",
" <td>-0.910792</td>\n",
" <td>0.064439</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.295759</td>\n",
" <td>0.323651</td>\n",
" <td>-0.027892</td>\n",
" <td>-0.090119</td>\n",
" <td>0.002514</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" current_dpd_base current_dpd_prod base-prod ln(base/prod) \\\n",
"(-inf, -30.0] 0.074499 0.037604 0.036895 0.683669 \n",
"(-30.0, -27.0] 0.081936 0.086814 -0.004878 -0.057830 \n",
"(-27.0, -24.0] 0.055914 0.042423 0.013491 0.276122 \n",
"(-24.0, -19.0] 0.080815 0.057078 0.023737 0.347743 \n",
"(-19.0, -15.0] 0.062045 0.048030 0.014016 0.256046 \n",
"(-15.0, -10.0] 0.077819 0.063307 0.014511 0.206382 \n",
"(-10.0, -6.0] 0.068961 0.069733 -0.000772 -0.011132 \n",
"(-6.0, -3.0] 0.064323 0.061275 0.003048 0.048545 \n",
"(-3.0, 0.0] 0.090326 0.091732 -0.001406 -0.015441 \n",
"(0.0, inf] 0.047602 0.118353 -0.070751 -0.910792 \n",
"NaN 0.295759 0.323651 -0.027892 -0.090119 \n",
"\n",
" index \n",
"(-inf, -30.0] 0.025224 \n",
"(-30.0, -27.0] 0.000282 \n",
"(-27.0, -24.0] 0.003725 \n",
"(-24.0, -19.0] 0.008254 \n",
"(-19.0, -15.0] 0.003589 \n",
"(-15.0, -10.0] 0.002995 \n",
"(-10.0, -6.0] 0.000009 \n",
"(-6.0, -3.0] 0.000148 \n",
"(-3.0, 0.0] 0.000022 \n",
"(0.0, inf] 0.064439 \n",
"NaN 0.002514 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"date_of_month PSI: 1.0104678350423415\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>date_of_month_base</th>\n",
" <th>date_of_month_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.047548</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0.044090</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>0.039356</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>0.029467</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>0.034424</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>0.034382</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>0.028610</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>0.032632</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>0.033585</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10</th>\n",
" <td>0.035401</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>11</th>\n",
" <td>0.031110</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12</th>\n",
" <td>0.033153</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>13</th>\n",
" <td>0.031877</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>14</th>\n",
" <td>0.027849</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>15</th>\n",
" <td>0.028407</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>16</th>\n",
" <td>0.028167</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>17</th>\n",
" <td>0.027142</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18</th>\n",
" <td>0.025194</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>19</th>\n",
" <td>0.027586</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20</th>\n",
" <td>0.025296</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>21</th>\n",
" <td>0.023960</td>\n",
" <td>0.108452</td>\n",
" <td>-0.084492</td>\n",
" <td>-1.509924</td>\n",
" <td>0.127577</td>\n",
" </tr>\n",
" <tr>\n",
" <th>22</th>\n",
" <td>0.021569</td>\n",
" <td>0.093273</td>\n",
" <td>-0.071704</td>\n",
" <td>-1.464282</td>\n",
" <td>0.104995</td>\n",
" </tr>\n",
" <tr>\n",
" <th>23</th>\n",
" <td>0.024913</td>\n",
" <td>0.080487</td>\n",
" <td>-0.055574</td>\n",
" <td>-1.172708</td>\n",
" <td>0.065172</td>\n",
" </tr>\n",
" <tr>\n",
" <th>24</th>\n",
" <td>0.023810</td>\n",
" <td>0.098715</td>\n",
" <td>-0.074905</td>\n",
" <td>-1.422126</td>\n",
" <td>0.106524</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25</th>\n",
" <td>0.038858</td>\n",
" <td>0.153433</td>\n",
" <td>-0.114574</td>\n",
" <td>-1.373335</td>\n",
" <td>0.157349</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26</th>\n",
" <td>0.039763</td>\n",
" <td>0.162809</td>\n",
" <td>-0.123046</td>\n",
" <td>-1.409631</td>\n",
" <td>0.173449</td>\n",
" </tr>\n",
" <tr>\n",
" <th>27</th>\n",
" <td>0.040261</td>\n",
" <td>0.172349</td>\n",
" <td>-0.132089</td>\n",
" <td>-1.454145</td>\n",
" <td>0.192076</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28</th>\n",
" <td>0.047692</td>\n",
" <td>0.130483</td>\n",
" <td>-0.082791</td>\n",
" <td>-1.006479</td>\n",
" <td>0.083328</td>\n",
" </tr>\n",
" <tr>\n",
" <th>29</th>\n",
" <td>0.033435</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30</th>\n",
" <td>0.034550</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>31</th>\n",
" <td>0.025914</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" date_of_month_base date_of_month_prod base-prod ln(base/prod) index\n",
"1 0.047548 NaN NaN NaN NaN\n",
"2 0.044090 NaN NaN NaN NaN\n",
"3 0.039356 NaN NaN NaN NaN\n",
"4 0.029467 NaN NaN NaN NaN\n",
"5 0.034424 NaN NaN NaN NaN\n",
"6 0.034382 NaN NaN NaN NaN\n",
"7 0.028610 NaN NaN NaN NaN\n",
"8 0.032632 NaN NaN NaN NaN\n",
"9 0.033585 NaN NaN NaN NaN\n",
"10 0.035401 NaN NaN NaN NaN\n",
"11 0.031110 NaN NaN NaN NaN\n",
"12 0.033153 NaN NaN NaN NaN\n",
"13 0.031877 NaN NaN NaN NaN\n",
"14 0.027849 NaN NaN NaN NaN\n",
"15 0.028407 NaN NaN NaN NaN\n",
"16 0.028167 NaN NaN NaN NaN\n",
"17 0.027142 NaN NaN NaN NaN\n",
"18 0.025194 NaN NaN NaN NaN\n",
"19 0.027586 NaN NaN NaN NaN\n",
"20 0.025296 NaN NaN NaN NaN\n",
"21 0.023960 0.108452 -0.084492 -1.509924 0.127577\n",
"22 0.021569 0.093273 -0.071704 -1.464282 0.104995\n",
"23 0.024913 0.080487 -0.055574 -1.172708 0.065172\n",
"24 0.023810 0.098715 -0.074905 -1.422126 0.106524\n",
"25 0.038858 0.153433 -0.114574 -1.373335 0.157349\n",
"26 0.039763 0.162809 -0.123046 -1.409631 0.173449\n",
"27 0.040261 0.172349 -0.132089 -1.454145 0.192076\n",
"28 0.047692 0.130483 -0.082791 -1.006479 0.083328\n",
"29 0.033435 NaN NaN NaN NaN\n",
"30 0.034550 NaN NaN NaN NaN\n",
"31 0.025914 NaN NaN NaN NaN"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"delin_max_dpd_90 PSI: 0.35325109478455907\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>delin_max_dpd_90_base</th>\n",
" <th>delin_max_dpd_90_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, -14.0]</th>\n",
" <td>0.088055</td>\n",
" <td>0.088879</td>\n",
" <td>-0.000825</td>\n",
" <td>-0.009321</td>\n",
" <td>0.000008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-14.0, -4.0]</th>\n",
" <td>0.097338</td>\n",
" <td>0.057636</td>\n",
" <td>0.039702</td>\n",
" <td>0.524049</td>\n",
" <td>0.020806</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-4.0, -2.0]</th>\n",
" <td>0.079191</td>\n",
" <td>0.039145</td>\n",
" <td>0.040046</td>\n",
" <td>0.704593</td>\n",
" <td>0.028216</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-2.0, -1.0]</th>\n",
" <td>0.084855</td>\n",
" <td>0.041473</td>\n",
" <td>0.043382</td>\n",
" <td>0.715903</td>\n",
" <td>0.031057</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-1.0, 0.0]</th>\n",
" <td>0.190714</td>\n",
" <td>0.122680</td>\n",
" <td>0.068034</td>\n",
" <td>0.441194</td>\n",
" <td>0.030016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0, 1.0]</th>\n",
" <td>0.074463</td>\n",
" <td>0.051341</td>\n",
" <td>0.023122</td>\n",
" <td>0.371811</td>\n",
" <td>0.008597</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 3.0]</th>\n",
" <td>0.067169</td>\n",
" <td>0.076093</td>\n",
" <td>-0.008924</td>\n",
" <td>-0.124745</td>\n",
" <td>0.001113</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.0, 9.0]</th>\n",
" <td>0.078448</td>\n",
" <td>0.124549</td>\n",
" <td>-0.046101</td>\n",
" <td>-0.462264</td>\n",
" <td>0.021311</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(9.0, inf]</th>\n",
" <td>0.082415</td>\n",
" <td>0.262639</td>\n",
" <td>-0.180223</td>\n",
" <td>-1.159006</td>\n",
" <td>0.208880</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.157352</td>\n",
" <td>0.135565</td>\n",
" <td>0.021787</td>\n",
" <td>0.149032</td>\n",
" <td>0.003247</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" delin_max_dpd_90_base delin_max_dpd_90_prod base-prod \\\n",
"(-inf, -14.0] 0.088055 0.088879 -0.000825 \n",
"(-14.0, -4.0] 0.097338 0.057636 0.039702 \n",
"(-4.0, -2.0] 0.079191 0.039145 0.040046 \n",
"(-2.0, -1.0] 0.084855 0.041473 0.043382 \n",
"(-1.0, 0.0] 0.190714 0.122680 0.068034 \n",
"(0.0, 1.0] 0.074463 0.051341 0.023122 \n",
"(1.0, 3.0] 0.067169 0.076093 -0.008924 \n",
"(3.0, 9.0] 0.078448 0.124549 -0.046101 \n",
"(9.0, inf] 0.082415 0.262639 -0.180223 \n",
"NaN 0.157352 0.135565 0.021787 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, -14.0] -0.009321 0.000008 \n",
"(-14.0, -4.0] 0.524049 0.020806 \n",
"(-4.0, -2.0] 0.704593 0.028216 \n",
"(-2.0, -1.0] 0.715903 0.031057 \n",
"(-1.0, 0.0] 0.441194 0.030016 \n",
"(0.0, 1.0] 0.371811 0.008597 \n",
"(1.0, 3.0] -0.124745 0.001113 \n",
"(3.0, 9.0] -0.462264 0.021311 \n",
"(9.0, inf] -1.159006 0.208880 \n",
"NaN 0.149032 0.003247 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEpCAYAAACduunJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3de5gU1Z3/8feXO16CRMhGBRkQiAKDoNxcES8RRBGQKBGzCHhDMURjNsniJlGWxH1Qyc9cvOJPjXGzwYgYUMAroCYiDKCCoAgI0REUAlFUHMIw3/2jaoaepmemkenqmprP63nmcbqquuszw/jt06dOnWPujoiIJFeDfAcQEZHcUqEXEUk4FXoRkYRToRcRSTgVehGRhGuU7wDpWrVq5QUFBfmOISJSpyxfvvzv7t46077YFfqCggKWLVuW7xgiInWKmf2tqn3quhERSTgVehGRhFOhFxFJuNj10WeyZ88eiouLKSkpyXcUqSeaNWtGmzZtaNy4cb6jiBy0OlHoi4uLOfzwwykoKMDM8h1HEs7d2b59O8XFxbRv3z7fcUQOWp3ouikpKeHII49UkZdImBlHHnmkPkFKYtSJQg+oyEuk9PcmSVJnCr2IiHw5daKPPl3BpLm1+nqbpg6p1dcTEYkTtei/pMmTJzNt2rSs9t900008//zzUUU7aIsWLeL8888/oOeMGzeOmTNnVrl/wYIFnHTSSXTr1o2xY8dSWloKBBc+r7vuOjp27Ej37t1ZsWJFreYSqdHkFtV/JYAKfQSmTJnC2Wefne8YeVNWVsbYsWOZMWMGb775Ju3atePhhx8GYP78+axbt45169Yxffp0JkyYkOe0IsmjQn8AbrnlFr7xjW9w9tlns3btWgA2bNjA4MGDOfnkkznttNN4++2393teamu3oKCAm2++mZNOOonCwsKMx5ebPHkyY8eOZdCgQRQUFDBr1ix+/OMfU1hYyODBg9mzZw8QvJH07t2bbt26MX78eNyd0tJSevfuzaJFiwC48cYb+clPflLluZ5++mmOP/54+vfvz6xZsypluPTSSznrrLPo1KkT999/PxC0xCdOnEiXLl0YMmQIW7durfK1t2/fTtOmTencuTMAAwcO5PHHHwdg9uzZjBkzBjOjX79+fPzxx2zZsqXK19q5cycjRoygS5cuXHPNNZSVlQEwYcIEevXqRdeuXbn55psrjp80aRJdunShe/fu/PCHPwRg27ZtXHjhhfTu3ZvevXvz17/+tcrziSSBCn2Wli9fzowZM3jttdeYNWsWRUVFAIwfP57f/va3LF++nGnTpnHttdfW+FqtWrVixYoVTJgwodruHwjeSObOncvs2bMZPXo0Z555JqtWraJ58+bMnRtcq5g4cSJFRUW8+eabfPHFFzz11FM0atSI3/3ud0yYMIHnnnuOp59+ulIBTFVSUsJVV13Fk08+ycsvv8yHH35Yaf/KlSuZO3cuixcvZsqUKWzevJknnniCtWvXsmrVKu6//35eeeWVan/ePXv2VExWN3PmTN5//30APvjgA9q2bVtxbJs2bfjggw+qfK2lS5fyy1/+klWrVrFhw4aKN6VbbrmFZcuWsXLlSl588UVWrlzJjh07eOKJJ1i9ejUrV67kpz/9KQDXX389N9xwA0VFRTz++ONceeWV1f4biNR1dfJibD68/PLLjBgxgkMOOQSAYcOGUVJSwiuvvMLIkSMrjtu9e3eNr/Wtb30LgJNPPrlS6zmTc889l8aNG1NYWMjevXsZPHgwAIWFhWzatAmAhQsXctttt7Fr1y527NhB165dGTp0KF27duXSSy9l6NChLF68mCZNmmQ8x9tvv0379u3p1KkTAKNHj2b69OkV+4cPH07z5s1p3rw5Z555JkuXLuWll17ikksuoWHDhhx99NGcddZZVf4MZsaMGTO44YYb2L17N4MGDaJRo+BPL9Pi9NUNbezTpw8dOnQA4JJLLuEvf/kLF110EX/605+YPn06paWlbNmyhTVr1tClSxeaNWvGlVdeyZAhQyr6959//nnWrFlT8Zo7d+7k008/5fDDD6/yvCJ1mQr9AUgvQGVlZRxxxBG8/vrrB/Q6TZs2BaBhw4YVFyVrOrZBgwY0bty4IkODBg0oLS2lpKSEa6+9lmXLltG2bVsmT55c6UafVatWccQRR/DRRx8d0M9W3b7yxwcy1vyUU07h5ZdfBuDZZ5/lnXfeAYIWfHnrHoK7oI8++ugDyrJx40amTZtGUVERLVu2ZNy4cZSUlNCoUSOWLl3KCy+8wIwZM7jzzjtZsGABZWVlLF68mObNm2edX6Quq5OFPh/DIQcMGMC4ceOYNGkSpaWlPPnkk1x99dW0b9+exx57jJEjR+LurFy5khNPPDGyXOVFvVWrVnz22WfMnDmTiy66CIBZs2axfft2XnrpJc4//3yWLl3KEUccsd9rHH/88WzcuJENGzZw3HHH8cc//rHS/tmzZ3PjjTfy+eefs2jRIqZOncrevXu57777GDNmDFu3bmXhwoV85zvfqTLn1q1b+drXvsbu3bu59dZbK64XDBs2jDvvvJNRo0axZMkSWrRowVFHHVXl6yxdupSNGzfSrl07Hn30UcaPH8/OnTs59NBDadGiBR999BHz58/njDPO4LPPPmPXrl2cd9559OvXj44dOwIwaNAg7rzzTn70ox8B8Prrr9OjR48D+K2L1C1ZFXozGwz8GmgI/H93n5q2/wfAlUApsA243N3/Fu7bC6wKD33P3YfVUvZInXTSSVx88cX06NGDdu3acdpppwHwhz/8gQkTJvCLX/yCPXv2MGrUqEgL/RFHHMFVV11FYWEhBQUF9O7dG4C///3vTJo0iRdeeIG2bdsyceJErr/++orRLqmaNWvG9OnTGTJkCK1ataJ///68+eabFfv79OnDkCFDeO+99/jZz37G0UcfzYgRI1iwYAGFhYV07tyZ008/vdqct99+O0899RRlZWVMmDChoqvnvPPOY968eXTs2JFDDjmEhx56qNrXOeWUU5g0aRKrVq1iwIABjBgxggYNGtCzZ0+6du1Khw4dOPXUUwH49NNPGT58OCUlJbg7d9xxBwC/+c1v+O53v0v37t0pLS1lwIAB3Hvvvdn/0kXqGMvUR1rpALOGwDvAQKAYKAIucfc1KcecCSxx911mNgE4w90vDvd95u6HZRuoV69enr7C1FtvvcUJJ5yQ7UtILZo8eTKHHXZYxYiV+kR/d/VETWPlJ38STY6DZGbL3b1Xpn3ZjLrpA6x393fd/Z/ADGB46gHuvtDdd4UPXwXaHExgERGpPdl03RwDvJ/yuBjoW83xVwDzUx43M7NlBN06U939z+lPMLPxwHiAY489NotIyfLQQw/x61//utK2U089lbvuuqvWzzVixAg2btxYadutt97KOeeck/H4yZMn5/T1M1m1ahWXXnpppW1NmzZlyZIlB5RFRALZFPpMQysy9veY2WigF5DaYXusu282sw7AAjNb5e4bKr2Y+3RgOgRdN1klT5DLLruMyy67LJJzPfHEE7F//cLCwgMeySQiVcum66YYaJvyuA2wOf0gMzsb+AkwzN0rBpO7++bwv+8Ci4CeB5FXREQOUDaFvgjoZGbtzawJMAqYk3qAmfUE7iMo8ltTtrc0s6bh962AU4E1iIhIZGrsunH3UjObCDxDMLzyQXdfbWZTgGXuPge4HTgMeCy8oaV8GOUJwH1mVkbwpjI1dbSOiIjkXlbj6N19HjAvbdtNKd9nnJrR3V8BCg8mYEa1PXVoHRk+JSLyZWhSsy9J89FXFsV89LVN89tLfaFCHwHNRx/tfPR79+496NcQSRIV+gOg+ejzOx/9okWLKqY9SJ+P/rDDDuOmm26ib9++LF68mBdeeIGePXtSWFjI5ZdfXjGraFU/p0iSqdBnSfPRx3s++s8//5xu3bqxZMkSevXqxbhx43j00UdZtWoVpaWl3HPPPTX+nCJJpUKfpdT56L/yla/sNx99jx49uPrqq6tdHalc6nz05XPKVyXb+ej79u1LYWEhCxYsYPXq1QCV5qN/8MEHs5qP3swYPXp0pf3l89G3atXqoOej79OnD4cffvhBz0ffsGHDivnoIZjy+cILLwRg7dq1tG/fvuITxNixY3nppZdq/DlFkqpOTlOcL5qPvvLjuMxHD8EMnA0bNgQyv3lU9XyR+qButugnf1K7X1kYMGAATzzxBF988QWffvopTz75JIccckjFfPQQFJg33ngjlz/5fjLNR18udT766667jo8//jjja6TORw9knI++pKSE7du3s2jRInr37s2AAQOYMWMGe/fuZcuWLSxcuLDanOV9+OXz0V9zzTVAMB/973//e9ydV199Nev56MvKynj00Ufp379/xp9n06ZNrF+/HoBHHnmE008/vcafUySp6mahz4PU+egvvPDCSvPRP/DAA5x44ol07dqV2bNnR5ordT76Cy64YL/56B944AE6d+5cMR99Jqnz0ffv35927dpV2l8+H32/fv0qzUffqVMnCgsLmTBhQlbz0Z9wwgl0796doUOHVpqPvkOHDnTs2JGrrrqKu+++u9rXKZ+Pvlu3brRv354RI0Zk/HkeeughRo4cSWFhIQ0aNOCaa66p8ecUSaoa56OPmuajj5c4zUe/aNEipk2bxlNPPRXJ+fR3V09oPnoREanrdDE2BjQffWXVzUd/xhlnHFAeEalDXTfHH3+8RkxIZNydt99+W1039YG6buKhWbNmbN++vdphcyK1xd3Zvn07zZo1y3cUkVpRJ7pu2rRpQ3FxMdu2bct3FKknmjVrRps2WvpYkqFOFPrGjRvTvn37fMcQEamT6kTXjYiIfHkq9CIiCadCLyKScCr0IiIJp0IvIpJwKvQiIgmnQi8iknAq9CIiCadCLyKScCr0IiIJp0IvIpJwKvQiIgmnQi8iknAq9CIiCadCLyKScFkVejMbbGZrzWy9mU3KsP8HZrbGzFaa2Qtm1i5l31gzWxd+ja3N8CIiUrMaC72ZNQTuAs4FugCXmFmXtMNeA3q5e3dgJnBb+NyvAjcDfYE+wM1m1rL24ouISE2yadH3Ada7+7vu/k9gBjA89QB3X+juu8KHrwLla7CdAzzn7jvc/R/Ac8Dg2okuIiLZyKbQHwO8n/K4ONxWlSuA+QfyXDMbb2bLzGyZ1oUVEald2RR6y7DNMx5oNhroBdx+IM919+nu3svde7Vu3TqLSCIikq1sCn0x0DblcRtgc/pBZnY28BNgmLvvPpDniohI7mRT6IuATmbW3syaAKOAOakHmFlP4D6CIr81ZdczwCAzaxlehB0UbhMRkYg0qukAdy81s4kEBboh8KC7rzazKcAyd59D0FVzGPCYmQG85+7D3H2Hmf2c4M0CYIq778jJTyIiIhnVWOgB3H0eMC9t200p359dzXMfBB78sgFFROTg6M5YEZGEy6pFL1IXFEyaW+3+TVOHRJREJF7UohcRSTgVehGRhFOhFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhFOhFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhFOhFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhFOhFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhGuU7wAiIklXMGlujcdsmjokZ+dXi15EJOGyKvRmNtjM1prZejOblGH/ADNbYWalZnZR2r69ZvZ6+DWntoKLiEh2auy6MbOGwF3AQKAYKDKzOe6+JuWw94BxwA8zvMQX7t6jFrKKiMiXkE0ffR9gvbu/C2BmM4DhQEWhd/dN4b6yHGQUEZGDkE3XzTHA+ymPi8Nt2WpmZsvM7FUzuyDTAWY2Pjxm2bZt2w7gpUVEpCbZFHrLsM0P4BzHunsv4DvAr8zsuP1ezH26u/dy916tW7c+gJcWEZGaZFPoi4G2KY/bAJuzPYG7bw7/+y6wCOh5APlEROQgZVPoi4BOZtbezJoAo4CsRs+YWUszaxp+3wo4lZS+fRERyb0aC727lwITgWeAt4A/uftqM5tiZsMAzKy3mRUDI4H7zGx1+PQTgGVm9gawEJiaNlpHRERyLKs7Y919HjAvbdtNKd8XEXTppD/vFaDwIDOKiMhB0J2xIiIJp0IvIpJwKvQiIgmnQi8iknAq9CIiCadCLyKScFp4REQkDia3qGH/J1/6pdWiFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhFOhFxFJOI2jF5H8yeHYcdlHLXoRkYRTi17qD7UepZ5Si15EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEi6rQm9mg81srZmtN7NJGfYPMLMVZlZqZhel7RtrZuvCr7G1FVxERLJTY6E3s4bAXcC5QBfgEjPrknbYe8A44H/TnvtV4GagL9AHuNnMWh58bBERyVY2Lfo+wHp3f9fd/wnMAIanHuDum9x9JVCW9txzgOfcfYe7/wN4DhhcC7lFRCRL2RT6Y4D3Ux4Xh9uykdVzzWy8mS0zs2Xbtm3L8qVFRCQb2RR6y7DNs3z9rJ7r7tPdvZe792rdunWWLy0iItnIZs3YYqBtyuM2wOYsX78YOCPtuYuyfG6sFUyaW+Mxm6YOiSCJiEj1smnRFwGdzKy9mTUBRgFzsnz9Z4BBZtYyvAg7KNwmIiIRqbHQu3spMJGgQL8F/MndV5vZFDMbBmBmvc2sGBgJ3Gdmq8Pn7gB+TvBmUQRMCbeJiEhEsum6wd3nAfPStt2U8n0RQbdMpuc+CDx4EBlFROQg6M5YEZGEU6EXEUk4FXoRkYTLqo8+bmoa2rip2XdqfpHJn9RSGhGReFOLXkQk4VToRUQSrk523Ui86C5hkXhTi15EJOHUoheRRKt58EZEQfJILXoRkYRTi76Oq7G1or5xkXpPhV5EciKri/T1oNskDtR1IyKScGrRi9RHk1vUsF93jieJCr1ILdI9BRJH6roREUk4FXoRkYRToRcRSTgVehGRhFOhFxFJOI26EUkgze8iqdSiFxFJOBV6EZGEU6EXEUk4FXoRkYRToRcRSTgVehGRhNPwylzSDIEiEgNq0YuIJJwKvYhIwmVV6M1ssJmtNbP1ZjYpw/6mZvZouH+JmRWE2wvM7Aszez38urd244uISE1q7KM3s4bAXcBAoBgoMrM57r4m5bArgH+4e0czGwXcClwc7tvg7j1qObeIiGQpm4uxfYD17v4ugJnNAIYDqYV+ODA5/H4mcKeZWS3mFEkOXaSXiGXTdXMM8H7K4+JwW8Zj3L0U+AQ4MtzX3sxeM7MXzey0TCcws/FmtszMlm3btu2AfgAREaleNoU+U8vcszxmC3Csu/cEfgD8r5l9Zb8D3ae7ey9379W6dessIomISLayKfTFQNuUx22AzVUdY2aNgBbADnff7e7bAdx9ObAB6HywoUVEJHvZFPoioJOZtTezJsAoYE7aMXOAseH3FwEL3N3NrHV4MRcz6wB0At6tnegiIpKNGi/GunupmU0EngEaAg+6+2ozmwIsc/c5wAPAI2a2HthB8GYAMACYYmalwF7gGnffkYsfREREMstqCgR3nwfMS9t2U8r3JcDIDM97HHj8IDOKiMhB0J2xIiIJp0nNJBoaOy6SN2rRi4gknFr0SVdTSxrUmhZJOLXoRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBJOhV5EJOFU6EVEEk6FXkQk4VToRUQSToVeRCThVOhFRBIuq0JvZoPNbK2ZrTezSRn2NzWzR8P9S8ysIGXfjeH2tWZ2Tu1FFxGRbNRY6M2sIXAXcC7QBbjEzLqkHXYF8A937wjcAdwaPrcLMAroCgwG7g5fT0REIpJNi74PsN7d33X3fwIzgOFpxwwHHg6/nwl808ws3D7D3Xe7+0Zgffh6IiISEXP36g8wuwgY7O5Xho8vBfq6+8SUY94MjykOH28A+gKTgVfd/X/C7Q8A8919Zto5xgPjw4ffANYe5M/VCvj7Qb5GbYhDjjhkgHjkiEMGiEeOOGSAeOSIQwY4+Bzt3L11ph2NsniyZdiW/u5Q1THZPBd3nw5MzyJLVsxsmbv3qq3Xq8s54pAhLjnikCEuOeKQIS454pAh1zmy6bopBtqmPG4DbK7qGDNrBLQAdmT5XBERyaFsCn0R0MnM2ptZE4KLq3PSjpkDjA2/vwhY4EGf0BxgVDgqpz3QCVhaO9FFRCQbNXbduHupmU0EngEaAg+6+2ozmwIsc/c5wAPAI2a2nqAlPyp87moz+xOwBigFvuvue3P0s6SqtW6ggxSHHHHIAPHIEYcMEI8cccgA8cgRhwyQwxw1XowVEZG6TXfGiogknAq9iEjCqdCLiCRcNuPoY83MfpDFYZ+7+305zvGtLA4rcfd5Oczw1SwOK3P3j3OVIcxxUhaH7XH3VTnMkPe/CzNbmcVh29z9m7nKEJccccgQ5thZ0yHAFnfvnMscUavzF2PNbAtwD5lvzir3b7n+hzOz7cDsGnIMcPfjcpihhOA+heoyNHT3Y3OVIczxKcGw3OpytHf3ghxmyPvfhZmtBs6r7hBgjrt3z1WGuOSIQ4Ywx2vu3vNgj6lr6nyLHnjE3adUd4CZHRpBjvnufnkNOf4nxxneyuaPOMcZAIrc/awacizIcYY4/F1c7e5/qyHDtTnOEJccccgA8J0sjrkw5ymo6AW4FfgawRudAe7uX6n1c9X1Fr3sY2bN3L3kYI+R2hd2q7m7/6O+58hnBjNb4e4nmdkj7n5p1OdPy7IeGOrub+X6XElo0RPOc38BcAzBXDqbgdnu/nTEOY4nmLEzNcecKP4hAdy9JJw1tE9ahqXhncpEVeTNrAXB1NSpOZ7J9fWBtAx5/bsws2OB24BvAh8Hm+wrwAJgkrtvqi854pAh1MTMxgL/mum6mrvPiigHwEdR1YY636I3s18BnYHfE8ytA8GcOmOAde5+fUQ5/gO4hGAa59Qcowimap4aQYZBwN3AOuCDlAwdgWvd/dlcZwhzjAFuBp5NyzEQ+C93/30EGfL+d2Fmi4FfATPL7wgP12MYCXzf3fvlOkNccsQhQ3jO/sC/Ad9m/6lcvKbu11rO8mvg68Cfgd0pIWr9zSYJhf6dTBfUwpbtO+7eKaocQFd335O2vQmwOoocZvYWcG566yicZ2ieu5+Q6wzh+dYSTGX9cdr2lsCSKEY0xOHvwszWVXWe6vYlMUccMqSd8wp3fyDKc2bI8FCGzTl5s0lC102JmfVx9/TJ0noDUfZFlwFHA+kXnI4K90WhEftar6k+ABpHlAHCi0oZtpdR/SiY2hSHv4vlZnY3waI874fb2hJMABjFRfE45YhDhgru/oCZ/StQQEodjOLTZsq5LovqXEko9OOAe8zscPYVubbAznBfVL4PvGBm69j3h3wsQbfJxCqfVbseBIrMbAaV/2caRTDxXFRuAVaY2bNU/l0MBH4eUYZx5P/vYgzBMpv/RXCdwAh+H08S7b9HHHLEIUMFM3sEOA54HSifaNEJuvqiytCM4HfSFWhWvj0XLfo633VTzsy+zr4/oGJ3/zAPGRqw70KoERSYoohm7CzPcAL7LgiXZ5jj7muiyhDmaAmck5bjmahHWsTh70LiJ+zm7OJ5LIBm9hjwNsGQzykE1w7eysX1o8QUepG6xMzOd/enlCM/GcIie527b4nyvGkZXnP3nma20t27m1ljgsZQtfegfBmJnuvGzFbkOwOAmcXhf+jJ+c4AYGZ5n/s7Jn8XvfMdIBSHHPnI0ApYY2bPmNmc8q+IM5QP3PjYzLoRrMxXkIsTqUUfATM7Kp8thzDDUHd/Mp8Zwhwnu/vyfOeQ+s3MTs+03d1fjDDDlcDjQCHwO+Aw4Ge5mH8pkYXezIaFK1/lO8dJ7h6H1mO9ZWb/QsoNU+7+UZ4jAWBmA939uQjP14dg6F6RmXUhuJnt7VxOspchw2HhedsSrDi3DnjW3aMalRYrZtaUYLqFAvaNivOapu74Uueq64U+w91tBtwFXAvR3elm+8/YaASTnA0l+D1HUvDDu0HbAC+kjqc3s8vd/cEoMlTHzKa7+/gIztMDuJfg43DqTVsfE9w8ltc3YDN7L9eTy6Wc62bgXIJRds8BfYFFwNkEfcK3RJDh28CPgDeAM4FXCLqOC4HR7p7N7Ja1keMv7t7fgon3UotfzuaZqSbL08AnwHL2jfzB3X9Z6+dKQKEvBZ4GtrJvjPZFwEwivNPNzMqAV0m5ww3oF27zXFxgyZDhv4H+wAqCN5hfuftvw30r3D2b6YNrI0dV0yUb8Ia7t4kgw+sEE2ktSdveD7jP3U+MIENVnyoNOMvdo5hsDzNbBfQAmgIfAm3cfaeZNSe4gS2nM0aGGVYC/dx9l5m1Av7g7ueYWXfgXnf/11xniBsze9Pdu0VxriSMoz8FmEowLe697u5mdkaUNyOEvg18D7i9/OOwmW109zMjzDAU6OnBgu6Tgf81sw7ufgPR3agEsI3gxrHUc3r4+GsRZTg0vcgDuPurFs1spgCnAaOBz9K2l89HFJXScIjvLjPb4O47Adz9i7CBEgUDvgi//5zw78DdV4Zz3tRHr5hZoedwXYZydb7Qh32OAwmK7IJwzpnIP6a4+8zwo9jPzewy4N/zkKORu5eGeT42s6HA9HAoWZMIc7wLfNPd30vfYWbvZzg+F+ab2VyCG2BSbx4bQ/AJMAqvArsyXeALp4mIyj/N7BB33wWcnJKhBdHdtT0PeNrMXiToRnoszPBVom2ExEl/YJyZbSToCSjvPqr1T1h1vusmlZkdTTBxUi9375DHHD2AO4Bu7t46wvM+RfCJ4sW07b8A/tPdIxlOa2bfBf7i7m9k2Pe98u6kCHKcS+abxyK7ABkHZtbU3Xdn2N4KOCqKFmV4vvOALgTdd8+F25sP0ZIAAAlDSURBVBoAjTPlSzoza5dpu9cwb/+XOleSCn2chJNnHV7+MTmiczaH4CN5hn3HuPsH+z9LRJIu6TdM5e3mnPDW6kURn/OL9CJffqNUvot8HG6UEqmKmT1vZvPN7Px8Z8mFOt9HX8MIj+rWqIxCHPoehwGT8x0C6JXvACLVGEMw02wk8+JHrc4XeuIxwqMqc/N8fojHmw0Ew19FYsNSljR0980EK5Al8q7tOt9HH04LXOUID3dvm4dYsWFmDerrnYeZWLAA9Xbg8fIRSnnI8DCwC7jL3d/MR4Ywx/ME863cla+JzaL+XViGJQ2BfCxpGKkk9NH/CmhZxb7bogwSRyry+zGCYW1Rrg2a7k7geSCvi1MTdFf8FMg4+iMiUf8uHgWeAL7u7p3cvSNBl82fCZYBTaQ636IXkeyldlfkO0s+WMyWNIxKElr0+9EID4Fg3h8zuyecgnZ2+P3gCM/fwsymmtnbZrY9/Hor3HZEhDmONbMZZrYNWEKwCtnWcFtBRBli8bsgXNLQzPqa2dHhV18LljmMfEnDqCSyRR/lvC41iUN/rJkNBz7MNCVAxDki6x83s18BnQnujC1fSrANQXfFulys4pMhwzMEfb8Pe7iylQUrXo0juK40MNcZwnMuJujinBlOhYCZNQRGAt9395yPNInR76IJwfJ9qTfSVSxpmNQbt5Ja6J9298habtUxs94E66X2cff/yFOG/yaYJbCRu5+bjwxhju8CxwPt3H1Yjs/1jrt3zrDdgHei+IhuZmvd/RsHui8HOfLeXRGX30V9lchCLxLOlniluy9N296HoOVWGEGGZwkuND7s4Tz4FsyPPw4Y6O5n5zpDeM4ZwA7gYSrP+zMWaOXu344gQyx+F9WxGCyrmCtJGEcfC+EEUTcCFwDl89tsJZiTfqq7fxxRjuPZ97HUCcYGz3H3t6I4f0qOcwh+F6k5Zrt7VBOKjQPuMbPD2dd10xbYGe6LwsXAJOBFMyu/p+MjYA7BbKdRGUPQXfFfZOiuiChDXH4X1ekNJLLQq0VfS+LQBxnO3HkJwTCx1H7pUcAMd5+a6wxhjrz3j6dk+Topk5qV/9uI1Ccq9LUkDn2QZvYO0NXd96RtbwKsjmroWBz6x+POYrLMZBy6K6L+XVg9XNIwkcMrIRjhYWYXm1lU3VN/M7Mfh/2O5Rn+JWxlRzUHexlwdIbtRxHdvOMAJWFfeLreQEmEOTIys7wXWGBCvgOEeuc7ABH+LixY0nAhQaGfSLAAzKXA6xasdpVIiW3RRznCIzxfS4I+yOHsm2OnvA/yVnffEUGGwQR3Gq5j35vLsUBHYGJU/eMWrJ97D5Cpf/xad0/kfCISf1ZPlzSs84XezK5391+b2anu/td858k3CxZy6EPlxTaKysdPR5yl3vePhxfpB1P5ovQzUV2cr4mZDSxfBCSCc30dwN0/NLPWBEstrnX31VGcP8ywCuju7m7B+g2vuHvPcF9ka7hGLQldN+Vrw0ayatGXEbZwI+HuZe7+qrs/7u4zw+8jL/Jhlg/dfbm7L4tTkQ//Z4/iPGMIFmo/AzgEOBQ4k+DuzDFRZMhCJKNuzOxqYDHwqplNIBjdcj4wy8yuiCJDqHxJw/8EnqWeLGmYhBb9HwkWCG8NbEjdRY7WXzxQZna/u1+V5wxPuXveF1WI6q5lM/tWVbsIPqLnfIlHC9aF7Zveeg+7+ZZkumCdoxxzqtoFnOXuOV8sPXxz7Qs0J5hWvGPYsm8JLHT3HrnOkJKl3i1pWOfH0bv7JeFHwmcIFtmInXwX+VAcMhDh1BSPAn8g8wLtzSLKYFWcv4xoW4+nAaOBz9K2G0E3XxT2eLA4+S4z21D+Cc/d/2FmkbU2zcw8WDO40rrB4Yib3SnH1O0WcJo6X+gh6CIATsx3jnRm9t/u/p95znCSu69w9y35zJEHK4FpmeYXMrOo7sK8BVgR3hWaenF8IPDziDIAvArs8rRF46HiU0cUysyscTj0d0jK+ZsRbRfyQjN7nODmvYo1LMIhyP0J7hZeCPwuwkw5V+e7bsqZ2akES+a1I3gDK++66RDR+X+Tvolg2NbvCYJcF0GG9NayEdyZO5Tg3zqSYYVm1ha4neAC5Hzg9vKx/Wb2Z3e/IIIMpwF/88wL0vRy92W5zhCeqyVwDpUvjj/j9WyaYAsW/NicPpmdmR0DnODuz0eUoxlwOfBvQHuCxUeaAQ0J+uzvcvfXo8gSpSQV+reBGwiWAqu4+Oju2yM6fzHBYuDPsu9j+TTgh2GOhyPIUEbQekvtZ+wXbnN3PyvXGcIczwGPh+e9AjgZGOru283stfJRDkmXTRdAFN0EccgRhwwZztcYaAV8EZdRULmShFE35T5x9/nuvtXdt5d/RXj+E4C/Ewylez4s7J+6+8NRFPnQtwmWhrvd3c909zMJpic+M6oiH2rt7ve6++vu/j3gbuAlMzuOzH3WkcjDjVILzex7YWs2NUcTMzvLgimsx9aTHHHIUIm773H3LUkv8pCsFv1Ugo9fs0hp0UZ9m7mZnUzQkp9LcJNSQcTnP4yg/7cN8O/Aoqi6r1IyrAZOdveSlG1nA/cCh7r7UVHmSckQ6aeJKroJmhM0sCLrJohDd0UcMtRnSSr0CzNsjqy7Ii2LAdcCp7j76KjPH2boAdwBdItiKGHauW8AVqRf/DOznsBtHtEiExly/cLdf5qnc8eimyAOOeKQob5JTKGX/YVvOIe7+858Z4laHPuERfKlzhd6Mxvt7v9jZj/ItN/d/1/UmdKZ2SqPYKGLKs4di2UVo85hZosILghXO4zO3X8XVSaRfEnCOPryu/oOz2eIGu7E/HqUWTKcPw6izjGYoE/4j2aWqX/8DvUJS31R51v0cWFme6j6TsyL3D0vb0T57JeOSw71CUt9l8hCn4/uCjNbDoyt4k7M9929bQQZYtEvHZccIhJI0jj6VPnorvg+wXzrmYyIKENcxirHJYeIkNwWfSy6K6IW83HbkecQkUCdL/Rx7ibI54iXuPRLxyWHSH2WhK6bOHcT5G3ES1xu745LDpH6LAnDK+M8jG5uns4rIlKhznfdpMpnN0Gcu5BEpH5LQtdNhTx3E8S5C0lE6rFEtejzSSNNRCSuVOhzQCNNRCROVOhFRBIuUX30IiKyPxV6EZGEU6EXEUk4FXoRkYT7P8nVJgC0L03wAAAAAElFTkSuQmCC\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"nondgtl_nonpl_trx_co_180 PSI: 0.29627296026825156\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>nondgtl_nonpl_trx_co_180_base</th>\n",
" <th>nondgtl_nonpl_trx_co_180_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.220961</td>\n",
" <td>0.281555</td>\n",
" <td>-0.060594</td>\n",
" <td>-0.242343</td>\n",
" <td>0.014685</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 2.0]</th>\n",
" <td>0.057616</td>\n",
" <td>0.073700</td>\n",
" <td>-0.016084</td>\n",
" <td>-0.246195</td>\n",
" <td>0.003960</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(2.0, 3.0]</th>\n",
" <td>0.036030</td>\n",
" <td>0.056226</td>\n",
" <td>-0.020196</td>\n",
" <td>-0.445030</td>\n",
" <td>0.008988</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.0, 6.0]</th>\n",
" <td>0.085712</td>\n",
" <td>0.105960</td>\n",
" <td>-0.020249</td>\n",
" <td>-0.212077</td>\n",
" <td>0.004294</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(6.0, 10.0]</th>\n",
" <td>0.065024</td>\n",
" <td>0.086060</td>\n",
" <td>-0.021036</td>\n",
" <td>-0.280291</td>\n",
" <td>0.005896</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(10.0, 18.0]</th>\n",
" <td>0.072659</td>\n",
" <td>0.081437</td>\n",
" <td>-0.008778</td>\n",
" <td>-0.114058</td>\n",
" <td>0.001001</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(18.0, 34.0]</th>\n",
" <td>0.069483</td>\n",
" <td>0.061307</td>\n",
" <td>0.008175</td>\n",
" <td>0.125174</td>\n",
" <td>0.001023</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(34.0, 84.0]</th>\n",
" <td>0.075763</td>\n",
" <td>0.031670</td>\n",
" <td>0.044093</td>\n",
" <td>0.872241</td>\n",
" <td>0.038460</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(84.0, inf]</th>\n",
" <td>0.075110</td>\n",
" <td>0.003672</td>\n",
" <td>0.071438</td>\n",
" <td>3.018247</td>\n",
" <td>0.215618</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.241643</td>\n",
" <td>0.218412</td>\n",
" <td>0.023231</td>\n",
" <td>0.101078</td>\n",
" <td>0.002348</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" nondgtl_nonpl_trx_co_180_base nondgtl_nonpl_trx_co_180_prod \\\n",
"(-inf, 1.0] 0.220961 0.281555 \n",
"(1.0, 2.0] 0.057616 0.073700 \n",
"(2.0, 3.0] 0.036030 0.056226 \n",
"(3.0, 6.0] 0.085712 0.105960 \n",
"(6.0, 10.0] 0.065024 0.086060 \n",
"(10.0, 18.0] 0.072659 0.081437 \n",
"(18.0, 34.0] 0.069483 0.061307 \n",
"(34.0, 84.0] 0.075763 0.031670 \n",
"(84.0, inf] 0.075110 0.003672 \n",
"NaN 0.241643 0.218412 \n",
"\n",
" base-prod ln(base/prod) index \n",
"(-inf, 1.0] -0.060594 -0.242343 0.014685 \n",
"(1.0, 2.0] -0.016084 -0.246195 0.003960 \n",
"(2.0, 3.0] -0.020196 -0.445030 0.008988 \n",
"(3.0, 6.0] -0.020249 -0.212077 0.004294 \n",
"(6.0, 10.0] -0.021036 -0.280291 0.005896 \n",
"(10.0, 18.0] -0.008778 -0.114058 0.001001 \n",
"(18.0, 34.0] 0.008175 0.125174 0.001023 \n",
"(34.0, 84.0] 0.044093 0.872241 0.038460 \n",
"(84.0, inf] 0.071438 3.018247 0.215618 \n",
"NaN 0.023231 0.101078 0.002348 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"oth_last_rep_days PSI: 0.04164111690499971\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>oth_last_rep_days_base</th>\n",
" <th>oth_last_rep_days_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.452242</td>\n",
" <td>0.517146</td>\n",
" <td>-0.064904</td>\n",
" <td>-0.134109</td>\n",
" <td>0.008704</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 2.0]</th>\n",
" <td>0.050916</td>\n",
" <td>0.055046</td>\n",
" <td>-0.004129</td>\n",
" <td>-0.077978</td>\n",
" <td>0.000322</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(2.0, 4.0]</th>\n",
" <td>0.066270</td>\n",
" <td>0.046095</td>\n",
" <td>0.020175</td>\n",
" <td>0.363030</td>\n",
" <td>0.007324</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(4.0, 8.0]</th>\n",
" <td>0.077573</td>\n",
" <td>0.046489</td>\n",
" <td>0.031084</td>\n",
" <td>0.512010</td>\n",
" <td>0.015915</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(8.0, 16.0]</th>\n",
" <td>0.079635</td>\n",
" <td>0.062357</td>\n",
" <td>0.017278</td>\n",
" <td>0.244581</td>\n",
" <td>0.004226</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(16.0, inf]</th>\n",
" <td>0.074307</td>\n",
" <td>0.091371</td>\n",
" <td>-0.017064</td>\n",
" <td>-0.206725</td>\n",
" <td>0.003528</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.199057</td>\n",
" <td>0.181496</td>\n",
" <td>0.017560</td>\n",
" <td>0.092354</td>\n",
" <td>0.001622</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" oth_last_rep_days_base oth_last_rep_days_prod base-prod \\\n",
"(-inf, 1.0] 0.452242 0.517146 -0.064904 \n",
"(1.0, 2.0] 0.050916 0.055046 -0.004129 \n",
"(2.0, 4.0] 0.066270 0.046095 0.020175 \n",
"(4.0, 8.0] 0.077573 0.046489 0.031084 \n",
"(8.0, 16.0] 0.079635 0.062357 0.017278 \n",
"(16.0, inf] 0.074307 0.091371 -0.017064 \n",
"NaN 0.199057 0.181496 0.017560 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, 1.0] -0.134109 0.008704 \n",
"(1.0, 2.0] -0.077978 0.000322 \n",
"(2.0, 4.0] 0.363030 0.007324 \n",
"(4.0, 8.0] 0.512010 0.015915 \n",
"(8.0, 16.0] 0.244581 0.004226 \n",
"(16.0, inf] -0.206725 0.003528 \n",
"NaN 0.092354 0.001622 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"oth_last_rep_dpd PSI: 0.12167329656894572\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>oth_last_rep_dpd_base</th>\n",
" <th>oth_last_rep_dpd_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, -35.0]</th>\n",
" <td>0.081660</td>\n",
" <td>0.080716</td>\n",
" <td>0.000944</td>\n",
" <td>0.011631</td>\n",
" <td>0.000011</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-35.0, -26.0]</th>\n",
" <td>0.083159</td>\n",
" <td>0.100059</td>\n",
" <td>-0.016900</td>\n",
" <td>-0.185011</td>\n",
" <td>0.003127</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-26.0, -19.0]</th>\n",
" <td>0.077951</td>\n",
" <td>0.067963</td>\n",
" <td>0.009988</td>\n",
" <td>0.137116</td>\n",
" <td>0.001369</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-19.0, -12.0]</th>\n",
" <td>0.086347</td>\n",
" <td>0.075077</td>\n",
" <td>0.011270</td>\n",
" <td>0.139857</td>\n",
" <td>0.001576</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-12.0, -7.0]</th>\n",
" <td>0.079982</td>\n",
" <td>0.063176</td>\n",
" <td>0.016806</td>\n",
" <td>0.235877</td>\n",
" <td>0.003964</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-7.0, -4.0]</th>\n",
" <td>0.077789</td>\n",
" <td>0.058488</td>\n",
" <td>0.019301</td>\n",
" <td>0.285177</td>\n",
" <td>0.005504</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-4.0, -2.0]</th>\n",
" <td>0.079707</td>\n",
" <td>0.059504</td>\n",
" <td>0.020202</td>\n",
" <td>0.292304</td>\n",
" <td>0.005905</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(-2.0, 0.0]</th>\n",
" <td>0.137988</td>\n",
" <td>0.110845</td>\n",
" <td>0.027143</td>\n",
" <td>0.219034</td>\n",
" <td>0.005945</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0, 1.0]</th>\n",
" <td>0.026010</td>\n",
" <td>0.029834</td>\n",
" <td>-0.003825</td>\n",
" <td>-0.137190</td>\n",
" <td>0.000525</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, inf]</th>\n",
" <td>0.070352</td>\n",
" <td>0.172841</td>\n",
" <td>-0.102490</td>\n",
" <td>-0.898868</td>\n",
" <td>0.092125</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.199057</td>\n",
" <td>0.181496</td>\n",
" <td>0.017560</td>\n",
" <td>0.092354</td>\n",
" <td>0.001622</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" oth_last_rep_dpd_base oth_last_rep_dpd_prod base-prod \\\n",
"(-inf, -35.0] 0.081660 0.080716 0.000944 \n",
"(-35.0, -26.0] 0.083159 0.100059 -0.016900 \n",
"(-26.0, -19.0] 0.077951 0.067963 0.009988 \n",
"(-19.0, -12.0] 0.086347 0.075077 0.011270 \n",
"(-12.0, -7.0] 0.079982 0.063176 0.016806 \n",
"(-7.0, -4.0] 0.077789 0.058488 0.019301 \n",
"(-4.0, -2.0] 0.079707 0.059504 0.020202 \n",
"(-2.0, 0.0] 0.137988 0.110845 0.027143 \n",
"(0.0, 1.0] 0.026010 0.029834 -0.003825 \n",
"(1.0, inf] 0.070352 0.172841 -0.102490 \n",
"NaN 0.199057 0.181496 0.017560 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, -35.0] 0.011631 0.000011 \n",
"(-35.0, -26.0] -0.185011 0.003127 \n",
"(-26.0, -19.0] 0.137116 0.001369 \n",
"(-19.0, -12.0] 0.139857 0.001576 \n",
"(-12.0, -7.0] 0.235877 0.003964 \n",
"(-7.0, -4.0] 0.285177 0.005504 \n",
"(-4.0, -2.0] 0.292304 0.005905 \n",
"(-2.0, 0.0] 0.219034 0.005945 \n",
"(0.0, 1.0] -0.137190 0.000525 \n",
"(1.0, inf] -0.898868 0.092125 \n",
"NaN 0.092354 0.001622 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"payment_type PSI: 0.0014869720266734998\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>payment_type_base</th>\n",
" <th>payment_type_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0.15683</td>\n",
" <td>0.171104</td>\n",
" <td>-0.014273</td>\n",
" <td>-0.087105</td>\n",
" <td>0.001243</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.84317</td>\n",
" <td>0.828896</td>\n",
" <td>0.014273</td>\n",
" <td>0.017073</td>\n",
" <td>0.000244</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" payment_type_base payment_type_prod base-prod ln(base/prod) index\n",
"0 0.15683 0.171104 -0.014273 -0.087105 0.001243\n",
"1 0.84317 0.828896 0.014273 0.017073 0.000244"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"pf_delin_max_dpd_12mo PSI: 0.052963641356337965\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>pf_delin_max_dpd_12mo_base</th>\n",
" <th>pf_delin_max_dpd_12mo_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 0.0]</th>\n",
" <td>0.380722</td>\n",
" <td>0.406924</td>\n",
" <td>-0.026202</td>\n",
" <td>-0.066558</td>\n",
" <td>0.001744</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0, 3.0]</th>\n",
" <td>0.020184</td>\n",
" <td>0.020851</td>\n",
" <td>-0.000667</td>\n",
" <td>-0.032499</td>\n",
" <td>0.000022</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.0, 19.0]</th>\n",
" <td>0.052852</td>\n",
" <td>0.081372</td>\n",
" <td>-0.028520</td>\n",
" <td>-0.431531</td>\n",
" <td>0.012307</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(19.0, 67.0]</th>\n",
" <td>0.055513</td>\n",
" <td>0.091994</td>\n",
" <td>-0.036481</td>\n",
" <td>-0.505107</td>\n",
" <td>0.018427</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(67.0, inf]</th>\n",
" <td>0.056508</td>\n",
" <td>0.053374</td>\n",
" <td>0.003134</td>\n",
" <td>0.057063</td>\n",
" <td>0.000179</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.434221</td>\n",
" <td>0.345486</td>\n",
" <td>0.088736</td>\n",
" <td>0.228603</td>\n",
" <td>0.020285</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" pf_delin_max_dpd_12mo_base pf_delin_max_dpd_12mo_prod \\\n",
"(-inf, 0.0] 0.380722 0.406924 \n",
"(0.0, 3.0] 0.020184 0.020851 \n",
"(3.0, 19.0] 0.052852 0.081372 \n",
"(19.0, 67.0] 0.055513 0.091994 \n",
"(67.0, inf] 0.056508 0.053374 \n",
"NaN 0.434221 0.345486 \n",
"\n",
" base-prod ln(base/prod) index \n",
"(-inf, 0.0] -0.026202 -0.066558 0.001744 \n",
"(0.0, 3.0] -0.000667 -0.032499 0.000022 \n",
"(3.0, 19.0] -0.028520 -0.431531 0.012307 \n",
"(19.0, 67.0] -0.036481 -0.505107 0.018427 \n",
"(67.0, inf] 0.003134 0.057063 0.000179 \n",
"NaN 0.088736 0.228603 0.020285 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEpCAYAAACDc9l6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3de5hU1Z3u8e8LgngLMYrGEQxowADSIraIMQJqiPoomkxAJWokxHglzsQJjudhTrwMzjGaoycmThLFiBIFxDFnOFFHZVC8DFEuchEIBA1qj4xRJDhGkdvv/FHVnaap7i6aqt5dq9/P8/jYtfeqqt+mqt/ee+2111ZEYGZmla9D1gWYmVlpONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBKxR1ZvfOCBB0bPnj2zenszs4q0cOHC9yKiW6F1mQV6z549WbBgQVZvb2ZWkSS90dg6d7mYmSXCgW5mlggHuplZIjLrQ7fmbdmyhZqaGjZt2pR1KWZl16VLF7p3706nTp2yLqViOdDbsJqaGvbbbz969uyJpKzLMSubiGD9+vXU1NTQq1evrMupWO5yacM2bdrEAQcc4DC35EnigAMO8NHobnKgt3EOc2sv/F3ffQ50M7NEuA+9gvS87rGSvt7aW85s8XOff/55Lr/8cjp16sS8efPYa6+9mmw/duxYzjrrLEaNGlVUm0suuYRrrrmGfv36tbjG1jRlyhQWLFjAT3/606KfM3z4cH70ox9RXV1dcP3EiRN54IEH2LBhAx9++GHd8ttvv53Jkyezxx570K1bN375y1/yuc99bre3oZBiPrdUtPT3a3d+j0rNe+jWIg8++CDf//73Wbx4cbNh3hKTJ0+umDAvl5EjR/Lyyy/vtPyYY45hwYIFLF26lFGjRnHttddmUJ21RQ50a9LatWv5whe+wMUXX0xVVRWjRo3izjvv5OGHH+amm27iggsuKPi8iGD8+PH069ePM888kz/+8Y916xYuXMiwYcM49thjOe2001i3bt1Ozx8+fHjd1BD77rsvEydO5Oijj2bIkCG88847jdY7duxYrrjiCk4++WQOP/xw5s6dy7hx4+jbty9jx46ta3fFFVdQXV1N//79uf766wHYuHEjRx55JKtWrQJgzJgx3HPPPY2+13333UefPn0YNmwYL7744g41XH755Zx00kn06dOH3/zmNwB8/PHHnH/++VRVVXHeeefx8ccfN/raAEOGDOGQQw7ZafnJJ5/M3nvvXdempqYGgGeffZZhw4Zx7rnn0qdPH6677joefPBBBg8ezIABA3jttdcAeOONNzj11FOpqqri1FNP5c0332yyjtmzZ++0LWvXruWkk05i0KBBDBo0iP/4j/8AYN26dQwdOpSBAwdy1FFH8fzzzwPw1FNPccIJJzBo0CBGjx69wxGHlY4D3Zq1atUqLr30UpYuXcqnPvUpNm/ezNlnn81tt93Ggw8+WPA5v/71r1m1ahXLli3jnnvuqfuF37JlC9/97nd55JFHWLhwIePGjWPixIlNvv+f//xnhgwZwpIlSxg6dGiTIQuwYcMG5syZwx133MHIkSP53ve+x/Lly1m2bBmLFy8G4Oabb67by507dy5Lly6la9eu/PSnP2Xs2LFMnz6dDRs28J3vfKfge6xbt47rr7+eF198kaeffpoVK1bssH7t2rXMnTuXxx57jMsvv5xNmzbxs5/9jL333pulS5cyceJEFi5c2OR2FOPee+/ljDPOqHu8ZMkSfvzjH7Ns2TKmTp3K6tWrefnll7nkkkv4yU9+AsD48eP55je/ydKlS7ngggu4+uqrm3yPQtty0EEH8fTTT7No0SJmzJhR9xoPPfQQp512GosXL2bJkiUMHDiQ9957j0mTJjF79mwWLVpEdXU1t99++25vu+3MfejWrB49enDiiScCcOGFF3LnnXfy6U9/usnnPPfcc4wZM4aOHTvyV3/1V5xyyilA7o/Dq6++yogRIwDYtm1bwb3Q+jp37sxZZ50FwLHHHsvTTz/dZPuRI0ciiQEDBnDwwQczYMAAAPr378/atWsZOHAgDz/8MHfffTdbt25l3bp1rFixgqqqKkaMGMHMmTO56qqrWLJkSaPv8dJLLzF8+HC6dctNenfeeeexevXquvXnnnsuHTp0oHfv3hx++OH87ne/47nnnqsLvqqqKqqqqprcjub86le/YsGCBcydO7du2XHHHVf373nEEUfwla98BYABAwbwzDPPADBv3jweffRRAC666KJmu2wKbUuvXr0YP348ixcvpmPHjnXbftxxxzFu3Di2bNnCV7/6VQYOHMjcuXNZsWJF3Xdo8+bNnHDCCbu17VaYA92a1XA4WbHDywq1iwj69+/PvHnzin7/Tp061b1Wx44d2bp1a5Pt99xzTwA6dOhQ93Pt461bt/KHP/yBH/3oR8yfP5/999+fsWPH1o1/3r59OytXrmSvvfbi/fffp3v37ru0fY2tq31cqqF5s2fP5uabb2bu3Lk7bGPD7a3/b9HYv1tzNRXaljvuuIODDz6YJUuWsH37drp06QLA0KFDee6553jssce46KKLmDBhAvvvvz8jRoxg2rRpLdpWK567XKxZb775Zl0AT5s2jS996UvNPmfo0KFMnz6dbdu2sW7durq9wyOPPJJ333237vW2bNnC8uXLy1d8AR988AH77LMPXbt25Z133uGJJ56oW3fHHXfQt29fpk2bVrenWcjxxx/Ps88+y/r169myZQszZ87cYf3MmTPZvn07r732Gq+//jpHHnkkQ4cOreuievXVV1m6dGmL6n/llVe47LLLmDVrFgcddNAuP/+LX/wi06dPB3Int5v7PAtty8aNGznkkEPo0KEDU6dOZdu2bUCuf/6ggw7iO9/5Dt/+9rdZtGgRQ4YM4cUXX2TNmjUAfPTRRzsczVjpeA+9gmQ1PKpv377cf//9XHbZZfTu3ZsrrriCK6+8ssnnfO1rX2POnDkMGDCg7sQh5LpPHnnkEa6++mo2btzI1q1b+du//Vv69+/fGpsCwNFHH80xxxxD//79Ofzww+u6AlavXs3kyZN5+eWX2W+//Rg6dCiTJk3ixhtv3Ok1DjnkEG644QZOOOEEDjnkEAYNGlQXapD7wzVs2DDeeecdfv7zn9OlSxeuuOIKvvWtb1FVVcXAgQMZPHhwk3Vee+21PPTQQ3z00Ud0796dSy65hBtuuIEJEybw4YcfMnr0aAAOO+wwZs2aVfT233nnnYwbN47bbruNbt26cd999zXZvtC2XHnllXz9619n5syZnHzyyeyzzz5A7sTsbbfdRqdOndh333154IEH6NatG1OmTGHMmDF88sknAEyaNIk+ffoUXbMVRxGRyRtXV1eHb3DRtJUrV9K3b99Ma1i7di1nnXUWr776aqZ1VJL2NHa71LL8zlfKOHRJCyOi4MUL7nIxM0uEu1ysST179mxy73zZsmVcdNFFOyzbc889eemll8pa180337xTv/Xo0aObHQLZEscff3xdV0GtqVOn1o2eaWjKlCllff1yaM1/Tysfd7m0YW2hy8WsNbnLpXnucjEzaweK6nKRdDrwY6AjMDkibmmk3ShgJnBcRLSt3e8burbweRtLW4eZWZk0u4cuqSNwF3AG0A8YI2mnWZMk7QdcDZS389TMzAoqpstlMLAmIl6PiM3AdOCcAu3+EbgV8C1HzMwyUEyXy6HAW/Ue1wDH128g6RigR0T8RtL3S1if1dfSbqNGX6/l3UmeD31Hqc6HXmoeo19exeyhF5rooW5ojKQOwB3A3zX7QtKlkhZIWvDuu+8WX6W1OZ4PvfwqZT705ubWsdZTTKDXAD3qPe4OvF3v8X7AUcCzktYCQ4BZknba7YiIuyOiOiKqa2eps7bN86G37/nQG9uWKVOmMHr0aEaOHMlXvvIVIoIJEyZw1FFHMWDAAGbMmAE0/T2w0ism0OcDvSX1ktQZOB+omzgiIjZGxIER0TMiegK/Bc5uc6NcrMU8H/rO2vt86JCbhvf+++9nzpw5PProo3VzoM+ePZsJEyawbt26Rr8HVh7NBnpEbAXGA08CK4GHI2K5pJsknV3uAi17DedDf+GFF5p9TjHzoQ8cOJBJkybV7WE2puF86GvXrm2yfaH50Dt06FA3HzrAww8/zKBBgzjmmGNYvnx5XSCPGDGCAQMGcNVVVzF58uRG36P+fOidO3fmvPPO22F9Y/OhX3jhhUBp50OfMGFC3bLa+dD33HPPneZDr932efPm8Y1vfAPIzYfe3OdZaFsg92/1mc98BoAXXnih7vM++OCDGTZsGPPnz2/0e2DlUdQ49Ih4HHi8wbIfNNJ2+O6XZW2J50MvfvsaW5fafOhA3QyLkPtci32+lY+vFLVmeT70nbX3+dAbGjp0KDNmzGDbtm28++67PPfccwwePLjR74GVhyfnqiQZXbXq+dA9H3rDbWnoa1/7GvPmzePoo49GErfeeiuf/exnG/0eWHm0n8m5KvDS/7YwOZfnQ991KY21bu1t8eRczfPkXGZm7UDFdbm0+K/ozkeJVgTPh96+50Pf1W2xbFVcl0vLA/0bLXpee+9yMWtN7nJpnrtcKlhWf3DNWpu/67vPgd6GdenShfXr1/uLbsmLCNavX19wBI0Vr+L60NuT7t27U1NTgycys/agS5cuTV7IZc1zoLdhnTp1olevXlmXYWYVwl0uZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlgjP5WJmtjva0O0tvYduZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpaIogJd0umSVklaI+m6Ausvl7RM0mJJL0jqV/pSzcysKc0GuqSOwF3AGUA/YEyBwH4oIgZExEDgVuD2kldqZmZNKmYPfTCwJiJej4jNwHTgnPoNIuKDeg/3AaJ0JZqZWTGKuUn0ocBb9R7XAMc3bCTpKuAaoDNwSqEXknQpcCnAYYcdtqu1mplZE4rZQ1eBZTvtgUfEXRFxBPD3wD8UeqGIuDsiqiOiulu3brtWqZmZNamYQK8BetR73B14u4n204Gv7k5RZma264oJ9PlAb0m9JHUGzgdm1W8gqXe9h2cCvy9diWZmVoxm+9AjYquk8cCTQEfglxGxXNJNwIKImAWMl/RlYAuwAbi4nEWbmdnOijkpSkQ8DjzeYNkP6v38NyWuy8zMdpGvFDUzS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEUUFuqTTJa2StEbSdQXWXyNphaSlkv5d0udKX6qZmTWl2UCX1BG4CzgD6AeMkdSvQbNXgOqIqAIeAW4tdaFmZta0YvbQBwNrIuL1iNgMTAfOqd8gIp6JiI/yD38LdC9tmWZm1pxiAv1Q4K16j2vyyxrzbeCJ3SnKzMx23R5FtFGBZVGwoXQhUA0Ma2T9pcClAIcddliRJZqZWTGK2UOvAXrUe9wdeLthI0lfBiYCZ0fEJ4VeKCLujojqiKju1q1bS+o1M7NGFBPo84HeknpJ6gycD8yq30DSMcAvyIX5H0tfppmZNafZQI+IrcB44ElgJfBwRCyXdJOks/PNbgP2BWZKWixpViMvZ2ZmZVJMHzoR8TjweINlP6j385dLXJeZme0iXylqZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFuZpYIB7qZWSIc6GZmiSgq0CWdLmmVpDWSriuwfqikRZK2ShpV+jLNzKw5zQa6pI7AXcAZQD9gjKR+DZq9CYwFHip1gWZmVpw9imgzGFgTEa8DSJoOnAOsqG0QEWvz67aXoUYzMytCMV0uhwJv1Xtck19mZmZtSDGBrgLLoiVvJulSSQskLXj33Xdb8hJmZtaIYgK9BuhR73F34O2WvFlE3B0R1RFR3a1bt5a8hJmZNaKYQJ8P9JbUS1Jn4HxgVnnLMjOzXdXsSdGI2CppPPAk0BH4ZUQsl3QTsCAiZkk6Dvg1sD8wUtKNEdG/rJWbpeKGri183sbS1mEVr5hRLkTE48DjDZb9oN7P88l1xZiZWUZ8paiZWSIc6GZmiXCgm5kloqg+dLPM+cShWbO8h25mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJ8FwuZlaUntc91qLnrb3lzBJXYo1xoJuVSIsDr0uJC7F2y4Furcqh1w55psxW4z50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLhQDczS4QD3cwsEQ50M7NEONDNzBLh6XNT4SlKzdo9B3ob4/nCzayl3OViZpYIB7qZWSIc6GZmiXCgm5klwoFuZpaIogJd0umSVklaI+m6Auv3lDQjv/4lST1LXaiZmTWt2UCX1BG4CzgD6AeMkdSvQbNvAxsi4vPAHcAPS12omZk1rZg99MHAmoh4PSI2A9OBcxq0OQe4P//zI8CpklS6Ms3MrDmKiKYbSKOA0yPikvzji4DjI2J8vTav5tvU5B+/lm/zXoPXuhS4NP/wSGBVqTakCAcC7zXbqnJ5+ypXytsG3r5S+1xEdCu0opgrRQvtaTf8K1BMGyLibuDuIt6z5CQtiIjqLN67NXj7KlfK2wbevtZUTJdLDdCj3uPuwNuNtZG0B9AVeL8UBZqZWXGKCfT5QG9JvSR1Bs4HZjVoMwu4OP/zKGBONNeXY2ZmJdVsl0tEbJU0HngS6Aj8MiKWS7oJWBARs4B7gamS1pDbMz+/nEW3UCZdPa3I21e5Ut428Pa1mmZPipqZWWXwlaJmZolwoJuZJcKBbmaWiCTvWCTpmiKa/TkiflH2YspAUsNRRoW8HxFjy11LqUkaVESzLRGxrOzFlEHKnx2ApL8uotmmiHi87MW0Q0meFJW0DvgZhS94qnVBRPRppZJKStLvgUuaagLcFRH9W6mkkpH03+SGyjb12fWKiJ6tU1FppfzZAUhaD/wrTX9+QyPiiFYqqV1Jcg8dmBoRNzXVQNI+rVVMGUyMiLlNNZB0Y2sVU2LzI+KUphpImtNaxZRByp8dwBMRMa6pBpJ+1VrFlEv+SOSHwEHk/ngJiIj4VKZ1pbiHbmbZkDQ6ImZK6hURf8i6nnLJX3MzMiJWZl1LfckGuqTTgK8Ch5KbV+Zt4F8j4t8yLawEJHUF/ge57audpOeP5A51b4mIP2VVWynkt+90dvzsnqz07YJ28dktiohBtf/Pup5ykfRiRJyYdR0NJRnokv4P0Ad4gNw8M5Cbg+abwO8j4m+yqq0UJD0JzAHuj4j/yi/7LLnpF74cESOyrG93SPomcD3wFPCf+cXdgRHAjRHxQFa1lULKnx2ApKfJdeUOBJ5vuD4izm71ospA0o+BzwL/F/ikdnlEPJpZUaQb6KsLnfDMz9G+OiJ6Z1BWyUhaFRFH7uq6SiBpFbmpl//UYPn+wEuVeiK7VsqfHUB+vqdBwFQKnPxt7vxBpZB0X4HF0dz5g3JL9aToJkmDI+LlBsuPAzZlUVCJvSHpWnJ7ee8ASDoYGAu8lWVhJSAKTL0MbKfpkROVIuXPjvxNcH4r6YsR8W7W9ZRLRHwr6xoKSTXQxwI/k7Qff+ly6QF8kF9X6c4DrgPmSjoov+wdcrNenptZVaVxM7BI0lP8JeAOI9fl8o+ZVVU6KX929e0v6WagJ/VyprkRTJVCUhdyt97sD3SpXZ71HnqSXS618n2Th5Lbs6up7bO0ti3fvXIa9T47cidFN2RamBVN0hLg58BCYFvt8ohYmFlRJSRpJvA74BvATcAFwMqsz88lHejtkaRBEbEo6zps16X02UlaGBHHZl1HuUh6JSKOkbQ0IqokdSK305HpEUi7m8tFUhK/ME24IusCykVSm5l3ukxS+uz+n6QrJR0i6TO1/2VdVAltyf//T5KOIneXtp7ZlZPjPXSrGJKOTeWQPXWSCl1UFBFxeKsXUwaSLgH+BRgATAH2Bf5n1vNDJR3o+dEDdRen1I4qSEH+/AAR8V+SugEnAasiYnm2lVlTJFVFxNKs67DdI2lP4Ovk9so75RdHc1OOlFuSgS5pILkTMl3Z8eKUPwFXVno/paTLyI2UELn5JMYCy4ETgVsj4t7sqts9kjqQ256vk/vMtgK/B34eEc9mV1lpSNoG/AGYBkyLiBUZl1RSkk6JiDmNzbqY9YU3pSLp34CN7HzS939nVhTpDlucAlwWES/VXyhpCHAfcHQWRZXQeHLDpfYC3gA+n99T3x94htw9XivVveS26X+Ru+H4B+SuOPwHSQMi4idZFlcCS4GLgDHALEl/Jhfu0yNibZaFlcgwclfCjiywLoAkAh3oHhGnZ11EQ6nuof++satBJa2JiM+3dk2lVH+eDElLIuLoeuteiYhjsqtu99SOGqj3+LcRMSR/iLs4IvpmWN5uazjHiaTB5G6qPhp4KyK+mFlxVrT8CfqftLV5+VPdQ39C0mPk5nKpvTilB7m5XCp+ci5gu6ROEbEFOLN2Yf5ih0ofubRF0hER8Vr+ZhebASLiE0kp7H3scLVr/mrmlyX9HTA0m5KsBb4EjM2f/P2Ev0yfW9X008oryUCPiKslnQGcw44Xp9yVyJ1S/pr85fERUVNv+QHA32VSUelMAJ6RtIncyabzAfInfn+TZWElcluhhZE7VE5inpN24oysCygkyS4Xq2z5SdQOiIj3sq6lNUg6ICLWZ12HVb5KPzy3BEXOTmEuqaKnlgWQdIukA/M/V0t6HXhJ0huShmVcXtnkt/XQrOtInffQrWJIejMiDsu6jt0haVlEDMj//AxwbUTMl9QHeCgiqrOtsDwk3Q9UkZu++rys60lVkn3oVrkkzWpsFblzBJWuk6Q9ImIrsFdEzAeIiNX5kTxJioiLAfIzoFqZtKs9dElXAuuBf8n/QiUlvxf0EbmTv69mXU9LSNoAXAh82HAVMCMiDm79qkpH0nfJjdG+hdyolk+TG5t9KnB4RFyUYXklkfItBNu69hboVwFfAD6Xyq2w6pN0HLm5wwdHxN9nXU9LSHqC3NWuzxRY91xEVPzQPknDyU3E1YfcUfJb5G5ldl9+KGrFSv0Wgm1dkoEu6W8i4seSToyIF7Oux6y9SP0Wgm1dqqNcam8PVemXiRckqWt+tMTvJK3P/7cyv+zTWddn7VrqtxBs01I9KbpS0lqgm6T6M9u1iau5SuBhcvNlDG9w5/ixwExyh7dmWUj9FoJtWpJdLlAXcE8CO/WVR8QbrV9R6TR1d/im1pm1Bt9CMDvJBnrK8ns/syl85/gREfHlDMuzFpBUDayLiP9strFZI1LtQwdA0omSnpa0WtLrkv6QvzKv0p1Hbkz2XEnvS3ofeBb4DGndOb6OpPsl/Sx/u68UfRf4jaQZWRdSLu3gFoKZS3oPXdLvgO+x8yT0njejwqQwJLMYkvaLiP/Ouo5y8C0Eyy/1QH8pIo7Puo7WpITuHJ8qX3hj5ZJ0lwu5aVhvk3SCpEG1/2VdVJlV9J3jUx+Smb/wZhEwHNgb2Ac4GViYX1fRJD0q6UJJ+2ZdS3uU+h76Tlcbkhu2eEqrF2NFkfQkuSGZ9xcYknlqRFT0kMzUL7yR9J/APOAUcifupwGPRcTmTAtrJ5IO9JSletie+pBMSauB4yJiY4PlXYEFjd06sVLU3gIxPwnXV8ndO/U4cjcnmRYRT2VaYOKSvLBI0oUR8StJ1xRaHxG3t3ZNpdTIfBknA/8kqdLny3hD0rUUHpL5VlNPrBCpX3hTeyet/wamAlMl1Y6+uo7cd9bKJMk9dEmXRcQvJF1faH1E3NjaNZVSyoft+W24jtztAw/KL34HmAX8MCLez6q2Ukn5wptUJlCrVEkGeupSP2w3s5ZJssulEEmLIiKVES6pH7YXlPqQTEl3R8SlWdexOyQdBvwxIjZJErmuskHACuCeFO9D0JakPmyxvmRmeouI+4FqcneJ/wTYTO5K0eqImJJdZWVX0UMyi/CLrAsogcf5S67cApwJvETuxKivFC2zdtPlImlSRPxD1nWUgiRFMx9cMW3MSk3Siojol/95Ibmuwe35x0si4uhMC0xcknvo+UO9HTQM80JtKsgzkr6bP7ytI6mzpFPyt6K7OKPadpukz+bHniOpm6S/ltQ/67pKQVIHSeMkPSZpiaSFkqYrdxejFLwlqfY6j7VADwBJKdwPts1Lcg9d0rPAvwD/GhFv1lveGfgSubB7plK7JyR1AcYBFwC9gD8Be5H7A/0UuXuKLs6uwpaTdBm5US4CfkiuD3Y5cCK5W9Pdm111u0/SfcAb5C66GQV8ADwP/D2572tF35RFUg/gAaAjsJHc79srwP7A9yPi3zMsL3mpBnqygdeQpE7AgcDHlX5REYCkZcDx5D6vN4DPR8R/5Yf6PRMRAzMtcDdJWlr/BiuSfhsRQyTtCSyOiL4Zllcykvryl3um1gDza7terHySHOUSEZuAfwb+ObXAayh/U+F1WddRQlsi4iPgI0mv1V7+HxEbJKWw97FF0hER8Vp+XqHNABHxSQrbV3vuJiJWAiubatPKpbULSQZ6fQkGXvvDc6IAAAEoSURBVOq2S+qU/9zOrF2YP+pK4ZzPBHLnQDYBnYDzIXeugNzl8ZXuGUnNdncCU7IpL21JdrlY5cqf6H274XhlSYcCfSNidjaVlU7+hPwBEfFe1rWUWiPdnV3I9akn1d3ZFjnQrU1pz0MyJY2IiKezrqNUUu/ubItSOIS1tCQ9JLMZFT2Cp6GI2BIR6xzmrcd76NampH7ILmlWY6uAUyJin9asx9LiQLc2K8VDdkkbgAuBDxuuAmZExMGtX5WlIvlRLla5Eh2h9Fvgo4iY23BFflpksxbzHrqZWSJ8UtSsFRUzh1CFzzNkGXKgm7Wu9jyKx8rMXS5mrSj1UTyWLQe6WUZSHMVj2XKgm5klwn3oZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJ+P8ffRhzdz4P9AAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"pl_settle_to_due_last_180 PSI: 0.08489189886876004\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>pl_settle_to_due_last_180_base</th>\n",
" <th>pl_settle_to_due_last_180_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.058396</td>\n",
" <td>0.061242</td>\n",
" <td>-0.002846</td>\n",
" <td>-0.047591</td>\n",
" <td>0.000135</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 5.0]</th>\n",
" <td>0.049017</td>\n",
" <td>0.029900</td>\n",
" <td>0.019117</td>\n",
" <td>0.494310</td>\n",
" <td>0.009450</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(5.0, 12.0]</th>\n",
" <td>0.052205</td>\n",
" <td>0.035408</td>\n",
" <td>0.016797</td>\n",
" <td>0.388251</td>\n",
" <td>0.006522</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(12.0, 20.0]</th>\n",
" <td>0.051989</td>\n",
" <td>0.036719</td>\n",
" <td>0.015270</td>\n",
" <td>0.347742</td>\n",
" <td>0.005310</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(20.0, 28.0]</th>\n",
" <td>0.052792</td>\n",
" <td>0.060422</td>\n",
" <td>-0.007630</td>\n",
" <td>-0.134996</td>\n",
" <td>0.001030</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(28.0, 40.0]</th>\n",
" <td>0.047500</td>\n",
" <td>0.046587</td>\n",
" <td>0.000913</td>\n",
" <td>0.019413</td>\n",
" <td>0.000018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(40.0, 60.0]</th>\n",
" <td>0.053577</td>\n",
" <td>0.049472</td>\n",
" <td>0.004105</td>\n",
" <td>0.079714</td>\n",
" <td>0.000327</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(60.0, 83.0]</th>\n",
" <td>0.048729</td>\n",
" <td>0.077700</td>\n",
" <td>-0.028971</td>\n",
" <td>-0.466581</td>\n",
" <td>0.013517</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(83.0, 118.0]</th>\n",
" <td>0.053038</td>\n",
" <td>0.100321</td>\n",
" <td>-0.047283</td>\n",
" <td>-0.637372</td>\n",
" <td>0.030137</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(118.0, inf]</th>\n",
" <td>0.050293</td>\n",
" <td>0.076847</td>\n",
" <td>-0.026554</td>\n",
" <td>-0.423955</td>\n",
" <td>0.011258</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.482465</td>\n",
" <td>0.425382</td>\n",
" <td>0.057083</td>\n",
" <td>0.125920</td>\n",
" <td>0.007188</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" pl_settle_to_due_last_180_base pl_settle_to_due_last_180_prod \\\n",
"(-inf, 1.0] 0.058396 0.061242 \n",
"(1.0, 5.0] 0.049017 0.029900 \n",
"(5.0, 12.0] 0.052205 0.035408 \n",
"(12.0, 20.0] 0.051989 0.036719 \n",
"(20.0, 28.0] 0.052792 0.060422 \n",
"(28.0, 40.0] 0.047500 0.046587 \n",
"(40.0, 60.0] 0.053577 0.049472 \n",
"(60.0, 83.0] 0.048729 0.077700 \n",
"(83.0, 118.0] 0.053038 0.100321 \n",
"(118.0, inf] 0.050293 0.076847 \n",
"NaN 0.482465 0.425382 \n",
"\n",
" base-prod ln(base/prod) index \n",
"(-inf, 1.0] -0.002846 -0.047591 0.000135 \n",
"(1.0, 5.0] 0.019117 0.494310 0.009450 \n",
"(5.0, 12.0] 0.016797 0.388251 0.006522 \n",
"(12.0, 20.0] 0.015270 0.347742 0.005310 \n",
"(20.0, 28.0] -0.007630 -0.134996 0.001030 \n",
"(28.0, 40.0] 0.000913 0.019413 0.000018 \n",
"(40.0, 60.0] 0.004105 0.079714 0.000327 \n",
"(60.0, 83.0] -0.028971 -0.466581 0.013517 \n",
"(83.0, 118.0] -0.047283 -0.637372 0.030137 \n",
"(118.0, inf] -0.026554 -0.423955 0.011258 \n",
"NaN 0.057083 0.125920 0.007188 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"pl_trx_suc_co_90 PSI: 0.03926826243109513\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>pl_trx_suc_co_90_base</th>\n",
" <th>pl_trx_suc_co_90_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.901943</td>\n",
" <td>0.952429</td>\n",
" <td>-0.050486</td>\n",
" <td>-0.054465</td>\n",
" <td>0.002750</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, inf]</th>\n",
" <td>0.098057</td>\n",
" <td>0.047571</td>\n",
" <td>0.050486</td>\n",
" <td>0.723334</td>\n",
" <td>0.036519</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" pl_trx_suc_co_90_base pl_trx_suc_co_90_prod base-prod \\\n",
"(-inf, 1.0] 0.901943 0.952429 -0.050486 \n",
"(1.0, inf] 0.098057 0.047571 0.050486 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, 1.0] -0.054465 0.002750 \n",
"(1.0, inf] 0.723334 0.036519 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"time_appr_to_pl_hour PSI: 0.08507020103153659\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>time_appr_to_pl_hour_base</th>\n",
" <th>time_appr_to_pl_hour_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 3.0]</th>\n",
" <td>0.102600</td>\n",
" <td>0.057504</td>\n",
" <td>0.045095</td>\n",
" <td>0.578974</td>\n",
" <td>0.026109</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.0, 65.0]</th>\n",
" <td>0.097847</td>\n",
" <td>0.077470</td>\n",
" <td>0.020377</td>\n",
" <td>0.233513</td>\n",
" <td>0.004758</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(65.0, 311.0]</th>\n",
" <td>0.099855</td>\n",
" <td>0.061898</td>\n",
" <td>0.037957</td>\n",
" <td>0.478238</td>\n",
" <td>0.018153</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(311.0, 852.0]</th>\n",
" <td>0.099777</td>\n",
" <td>0.094879</td>\n",
" <td>0.004898</td>\n",
" <td>0.050336</td>\n",
" <td>0.000247</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(852.0, 2154.0]</th>\n",
" <td>0.099927</td>\n",
" <td>0.093240</td>\n",
" <td>0.006687</td>\n",
" <td>0.069264</td>\n",
" <td>0.000463</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(2154.0, 3717.0]</th>\n",
" <td>0.100095</td>\n",
" <td>0.111173</td>\n",
" <td>-0.011078</td>\n",
" <td>-0.104971</td>\n",
" <td>0.001163</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3717.0, 5238.7]</th>\n",
" <td>0.099897</td>\n",
" <td>0.158481</td>\n",
" <td>-0.058584</td>\n",
" <td>-0.461498</td>\n",
" <td>0.027037</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(5238.7, 7565.0]</th>\n",
" <td>0.100023</td>\n",
" <td>0.107403</td>\n",
" <td>-0.007380</td>\n",
" <td>-0.071188</td>\n",
" <td>0.000525</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(7565.0, 11549.0]</th>\n",
" <td>0.099987</td>\n",
" <td>0.119927</td>\n",
" <td>-0.019940</td>\n",
" <td>-0.181841</td>\n",
" <td>0.003626</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(11549.0, inf]</th>\n",
" <td>0.099993</td>\n",
" <td>0.118025</td>\n",
" <td>-0.018032</td>\n",
" <td>-0.165799</td>\n",
" <td>0.002990</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" time_appr_to_pl_hour_base time_appr_to_pl_hour_prod \\\n",
"(-inf, 3.0] 0.102600 0.057504 \n",
"(3.0, 65.0] 0.097847 0.077470 \n",
"(65.0, 311.0] 0.099855 0.061898 \n",
"(311.0, 852.0] 0.099777 0.094879 \n",
"(852.0, 2154.0] 0.099927 0.093240 \n",
"(2154.0, 3717.0] 0.100095 0.111173 \n",
"(3717.0, 5238.7] 0.099897 0.158481 \n",
"(5238.7, 7565.0] 0.100023 0.107403 \n",
"(7565.0, 11549.0] 0.099987 0.119927 \n",
"(11549.0, inf] 0.099993 0.118025 \n",
"\n",
" base-prod ln(base/prod) index \n",
"(-inf, 3.0] 0.045095 0.578974 0.026109 \n",
"(3.0, 65.0] 0.020377 0.233513 0.004758 \n",
"(65.0, 311.0] 0.037957 0.478238 0.018153 \n",
"(311.0, 852.0] 0.004898 0.050336 0.000247 \n",
"(852.0, 2154.0] 0.006687 0.069264 0.000463 \n",
"(2154.0, 3717.0] -0.011078 -0.104971 0.001163 \n",
"(3717.0, 5238.7] -0.058584 -0.461498 0.027037 \n",
"(5238.7, 7565.0] -0.007380 -0.071188 0.000525 \n",
"(7565.0, 11549.0] -0.019940 -0.181841 0.003626 \n",
"(11549.0, inf] -0.018032 -0.165799 0.002990 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAFJCAYAAACVTuaTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3de7xUdb3/8ddbUFFRLOCUigqpnSMBKnLRvEulVoo3FC2NLofSo3ZOHX9SmRppaVpq3lIzb+UlzaOUGNpR7FheQG6KV1SULR6P4iUvkSDv3x/ftTfDMJs9W2bWzF58no8HD2bWWjPfz5695zNrvuv7/XxlmxBCCMW1VqMDCCGEUF+R6EMIoeAi0YcQQsFFog8hhIKLRB9CCAUXiT6EEAque6MDKNenTx/379+/0WGEEEKX8vDDD79qu2+lfU2X6Pv378/06dMbHUYIIXQpkp5vb1903YQQQsFFog8hhIKLRB9CCAVXVR+9pH2B84FuwC9tn1m2f3fgPGAIMNb2zSX7tgB+CWwOGPis7fmdCXLJkiW0tLSwePHizjwshNz16NGDfv36sfbaazc6lBDadJjoJXUDLgI+DbQA0yRNsv1YyWEvAOOA/6zwFNcAZ9i+S1JPYFlng2xpaWHDDTekf//+SOrsw0PIhW0WLVpES0sLAwYMaHQ4IbSpputmBDDP9rO23wNuAEaXHmB7vu05lCVxSQOB7rbvyo572/a7nQ1y8eLF9O7dO5J8aGqS6N27d3zzDE2nmkS/GbCg5H5Ltq0aHwfekHSLpJmSzs6+IaxA0nhJ0yVNf+WVVyo+UST50BXE32loRtUk+kp/udUWse8O7Ebq0hkOfIzUxbPik9mX2R5me1jfvhXH+4cQQviAqrkY20K6kNqqH7CwyudvAWbafhZA0q3ATsAVnQmyXP8Jt6/Ow1cy/8zPdXjMG2+8wXXXXcexxx7LwoULOeGEE7j55ps7fFxXcN555zF+/HjWX3/9mj3nuHHj+PznP8+hhx5acX/rxLg+ffrUrM1V6dmzJ2+//XYubXUJp/XqYP+b+cQRclHNGf00YBtJAyStA4wFJlX5/NOAD0lqPU3fG3hsFcc3rTfeeIOLL74YgE033bRLJvn333+/4vbzzjuPd9/t9KWTpmObZcs6fa0/hMLrMNHbXgocB0wBHgd+a3uupImSDgCQNFxSCzAGuFTS3Oyx75O6bf5b0iOkbqDL6/Oj1NeECRN45pln2H777RkzZgyDBg0C4KqrruLAAw9k//33Z8CAAVx44YX87Gc/Y4cddmCnnXbitddeA+CZZ55h3333Zccdd2S33XbjiSeeaLet3//+94wcOZIddtiBT33qU7z88ssAnHbaaRx11FHsvffebLPNNlx+eXopp06dyu67785BBx3EwIED+cY3vtGW8Hr27Mkpp5zCyJEjuf/++1dq6+c//zkLFy5kr732Yq+99gLg+uuvZ/DgwQwaNIiTTjppla9Lz549+fa3v83QoUMZNWoU7V1jqeSCCy5g6NChDB48uO31eO211zjwwAMZMmQIO+20E3PmzGn72c8555y2xw4aNIj58+czf/58tt12W4499liGDh3KggULKrYFVIzz8ssvZ/jw4Wy33XYccsghbR94N910E4MGDWK77bZj9913B9IH5Yknnsjw4cMZMmQIl156adU/awiNVNWEKduTbX/c9la2z8i2nWJ7UnZ7mu1+tjew3dv2J0oee5ftIbYH2x6Xjdzpcs4880y22morZs2axdlnn73CvkcffZTrrruOhx56iO9973usv/76zJw5k5133plrrrkGgPHjx3PBBRfw8MMPc84553Dssce229auu+7KAw88wMyZMxk7diw/+clP2vbNmTOH22+/nfvvv5+JEyeycGHqRXvooYf46U9/yiOPPMIzzzzDLbfcAsA777zDoEGDePDBB9l1111XauuEE05g00035Z577uGee+5h4cKFnHTSSdx9993MmjWLadOmceutt7Yb6zvvvMPQoUOZMWMGe+yxBz/4wQ+qfk379OnDjBkzOOaYY9qS+KmnnsoOO+zAnDlz+NGPfsTRRx/d4fM8+eSTHH300cycOZMtt9yyU3EefPDBTJs2jdmzZ7PttttyxRWpV3HixIlMmTKF2bNnM2lS+gJ7xRVX0KtXL6ZNm8a0adO4/PLLee6556r+eUNolKYratYV7bXXXmy44YZsuOGG9OrVi/333x+AwYMHM2fOHN5++23++te/MmbMmLbH/OMf/2j3+VpaWjj88MN56aWXeO+991YYkz169GjWW2891ltvPfbaay8eeughNt54Y0aMGMHHPvYxAI444gjuu+8+Dj30ULp168YhhxxS9c8ybdo09txzT1ovin/hC1/gz3/+MwceeGDF49daay0OP/xwAL74xS9y8MEHV91W67E77rhj2wfTfffdx+9+9zsA9t57bxYtWsSbb666v3jLLbdkp512WuUx7cX56KOPcvLJJ/PGG2/w9ttvs88++wCwyy67MG7cOA477LC2Y++8807mzJnT1m335ptv8vTTT8eY+dD0ItHXwLrrrtt2e6211mq7v9Zaa7F06VKWLVvGxhtvzKxZs6p6vuOPP55vfetbHHDAAUydOpXTTjutbV/58L3W++1t79GjB926rTSitV12tQOqKuvM8MLW16lbt24sXbq03fYl0b179xX630vHqm+wwQYfOM5x48Zx6623st1223HVVVcxdepUAH7xi1/w4IMPcvvtt7P99tsza9YsbHPBBRe0fRiE0FVErZsqbbjhhrz11lsf6LEbbbQRAwYM4KabbgJSMps9e3a7x7/55ptstlmaqnD11VevsO+2225j8eLFLFq0iKlTpzJ8+HAgdd0899xzLFu2jBtvvLFiN017Sn+2kSNHcu+99/Lqq6/y/vvvc/3117PHHnu0+9hly5a1neFed911nWq3kt13353f/OY3QLr20KdPHzbaaCP69+/PjBkzAJgxY0anu0zai/Ott95ik002YcmSJW3tQrqmMnLkSCZOnEifPn1YsGAB++yzD5dccglLliwB4KmnnuKdd95ZrZ83hDx0yTP6aoZD1lrv3r3ZZZddGDRoENtuu22nH/+b3/yGY445htNPP50lS5YwduxYtttuu4rHnnbaaYwZM4bNNtuMnXbaaYWkNmLECD73uc/xwgsv8P3vf59NN92Up556ip133pkJEybwyCOPtF2Yrdb48ePZb7/92GSTTbjnnnv48Y9/zF577YVtPvvZzzJ69Oh2H7vBBhswd+5cdtxxR3r16sWNN95Y/YvSzs/+5S9/mSFDhrD++uu3fdAdcsghXHPNNWy//fYMHz6cj3/845163vbi/OEPf8jIkSPZcsstGTx4cNsH3oknnsjTTz+NbUaNGsV2223HkCFDmD9/PkOHDsU2ffv2XeX1ixCahVb3q3qtDRs2zOULjzz++OMfKLkWzWmnnUbPnj35z/9csaTQ1KlTOeecc/jDH/6Qe0wxPn1lXeLvNcbRF46kh20Pq7Qvum5CCKHgumTXTVGcccYZbf32rcaMGcP3vve9iseXXpQtteeee7Lnnnt22N5BBx20Ut/2WWedVdXFxZEjR640Uujaa6/t8Gx+ddr8INqLc/DgwXVpL4SuILpuQqixLvH3Gl03hRNdNyGEsAaLRB9CCAUXiT6EEAouEn0IIRRc1xx109GFpE4/X8cXnqIefec0Wz36Wov5A6EriTP6KkU9+ua3uvXoW+vthFA0keirFPXoK+sq9ejbi3PPPffku9/9LnvssQfnn38+zz//PKNGjWLIkCGMGjWKF154AYDnnnuOnXfemeHDh/P973+/6p8xhGYQib5KUY++sq5ejx7St7V7772Xb3/72xx33HEcffTRzJkzhy984QuccMIJAHzzm9/kmGOOYdq0aXz0ox+t+mcMoRlEoq+B1nr0ffv2Xake/fz581eoR7/99tvz9a9/nZdeeqnd52tpaWGfffZh8ODBnH322cydO7dtX2s9+j59+rTVowfa6tF369atrR49sFr16Lt3795Wj7495XXeW9utRmk9+vnz5wOpHv1RRx0F1LcefWmcrdsB7r//fo488kgAjjrqqLbj/vKXv3DEEUe0bQ+hK6kq0UvaV9KTkuZJmlBh/+6SZkhaKmmlq2+SNpL0oqQLaxF0s+lMPfrWf48//ni7z3f88cdz3HHH8cgjj3DppZeuUHs96tEntapH39HjS4/rzM8WQjPpMNFL6gZcBOwHDASOkDSw7LAXgHHAde08zQ+Bez94mI0X9egr6+r16Mt98pOf5IYbbgBSaenW43bZZZcVtofQlVQzvHIEMM/2swCSbgBGA4+1HmB7frZvpSEPknYEPgL8EahYh6HTGlCHI+rRV9bV69GX+/nPf85XvvIVzj77bPr27cuVV14JwPnnn8+RRx7J+eef36musBCaQYdFzbKumH1tfy27fxQw0vZxFY69CviD7Zuz+2sBdwNHAaOAYZUeVyqKmrUv6tF/cHnG2SX+XqOoWeGsblGzSh2T1XbkHgtMtl15zFtrA9J4SdMlTe/M8LwQQggdq6brpgXYvOR+P2Bhlc+/M7CbpGOBnsA6kt62vcIFXduXAZdBOqOv8rm7vKhHX3sfNM4QiqyarpvuwFOkrpcXgWnAkbbnVjj2Kkq6bsr2jSO6bsIaoEv8vUbXTeGsVteN7aXAccAU4HHgt7bnSpoo6YCsgeGSWoAxwKWSVvoQWF3NtkBKCJXE32loRlUVNbM9GZhctu2UktvTSF06q3qOq4CrOh0haSz4okWL6N27d4xlDk3LNosWLaJHjx6NDiWEFXSJ6pX9+vWjpaWlU3VUQmiEHj160K/fKs95Qshdl0j0a6+9NgMGDGh0GCGE0CVFrZsQQii4SPQhhFBwkehDCKHgItGHEELBRaIPIYSCi0QfQggFF4k+hBAKLhJ9CCEUXCT6EEIouEj0IYRQcJHoQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCq6qRC9pX0lPSponaUKF/btLmiFpqaRDS7ZvL+l+SXMlzZF0eC2DDyGE0LEOE72kbsBFwH7AQOAISQPLDnsBGAdcV7b9XeBo258A9gXOk7Tx6gYdQgihetWsMDUCmGf7WQBJNwCjgcdaD7A9P9u3rPSBtp8qub1Q0v8BfYE3VjvyEEIIVamm62YzYEHJ/ZZsW6dIGgGsAzzT2ceGEEL44Ko5o1eFbe5MI5I2Aa4FvmR7WYX944HxAFtssUVnnjqEEFbPab062P9mPnHUUTVn9C3A5iX3+wELq21A0kbA7cDJth+odIzty2wPsz2sb9++1T51CCGEKlRzRj8N2EbSAOBFYCxwZDVPLmkd4L+Aa2zf9IGjDCGEoqvjN4sOz+htLwWOA6YAjwO/tT1X0kRJBwBIGi6pBRgDXCppbvbww4DdgXGSZmX/tv/A0YYQQui0as7osT0ZmFy27ZSS29NIXTrlj/s18OvVjDGEEMJqiJmxIYRQcJHoQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCq6q4ZUhhBpaA6bch+YSZ/QhhFBwkehDCKHgItGHEELBRaIPIYSCi4uxIYTGiQvTuYgz+hBCKLhI9CGEUHCR6EMIoeAi0YcQQsFFog8hhIKratSNpH2B84FuwC9tn1m2f3fgPGAIMNb2zSX7vgScnN093fbVqxt0/wm3r3L//DM/t7pNhBBCYXR4Ri+pG3ARsB8wEDhC0sCyw14AxgHXlT32w8CpwEhgBHCqpA+tftghhBCqVU3XzQhgnu1nbb8H3ACMLj3A9nzbc4BlZY/dB7jL9mu2XwfuAvatQdwhhBCqVE2i3wxYUHK/JdtWjdV5bAghhBqopo9eFba5yuev6rGSxgPjAbbYYosqnzpAc1yv6CiGZomjGWIAmN+j7mF0/Fo0QQzNEkczxFDvOKo5o28BNi+53w9YWOXzV/VY25fZHmZ7WN++fat86hBCCNWoJtFPA7aRNEDSOsBYYFKVzz8F+IykD2UXYT+TbQshhJCTDrtubC+VdBwpQXcDfmV7rqSJwHTbkyQNB/4L+BCwv6Qf2P6E7dck/ZD0YQEw0fZrdfpZctUs3RUhhNCRqsbR254MTC7bdkrJ7WmkbplKj/0V8KvViDGEEMJqiJmxIYRQcJHoQwih4GLhkZCPWGAihIaJM/oQQii4OKMPa474VhHWUHFGH0IIBReJPoQQCi4SfQghFFwk+hBCKLhI9CGEUHCR6EMIoeAi0YcQQsFFog8hhIKLRB9CCAUXiT6EEAouEn0IIRRcJPoQQii4qhK9pH0lPSlpnqQJFfavK+nGbP+Dkvpn29eWdLWkRyQ9Luk7tQ0/hBBCRzpM9JK6ARcB+wEDgSMkDSw77KvA67a3Bs4Fzsq2jwHWtT0Y2BH4euuHQAghhHxUc0Y/Aphn+1nb7wE3AKPLjhkNXJ3dvhkYJUmAgQ0kdQfWA94D/laTyEMIIVSlmkS/GbCg5H5Ltq3iMbaXAm8CvUlJ/x3gJeAF4Bzbr61mzCGEEDqhmkSvCttc5TEjgPeBTYEBwLclfWylBqTxkqZLmv7KK69UEVIIIYRqVZPoW4DNS+73Axa2d0zWTdMLeA04Evij7SW2/w/4CzCsvAHbl9keZntY3759O/9ThBBCaFc1iX4asI2kAZLWAcYCk8qOmQR8Kbt9KHC3bZO6a/ZWsgGwE/BEbUIPIYRQjQ7XjLW9VNJxwBSgG/Ar23MlTQSm254EXAFcK2ke6Ux+bPbwi4ArgUdJ3TtX2p5Th5+jOcUapSGEJlDV4uC2JwOTy7adUnJ7MWkoZfnj3q60PYQQQn5iZmwIIRRcJPoQQii4SPQhhFBwkehDCKHgItGHEELBRaIPIYSCi0QfQggFF4k+hBAKLhJ9CCEUXCT6EEIouEj0IYRQcJHoQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCi4SfQghFFxViV7SvpKelDRP0oQK+9eVdGO2/0FJ/Uv2DZF0v6S5kh6R1KN24YcQQuhIh4leUjfS2q/7AQOBIyQNLDvsq8DrtrcGzgXOyh7bHfg18A3bnwD2BJbULPoQQggdqmbN2BHAPNvPAki6ARgNPFZyzGjgtOz2zcCFkgR8BphjezaA7UU1ijtUq6MFyiEWKQ+h4KrputkMWFByvyXbVvEY20uBN4HewMcBS5oiaYak/7f6IYcQQuiMas7oVWGbqzymO7ArMBx4F/hvSQ/b/u8VHiyNB8YDbLHFFlWEFEIIoVrVnNG3AJuX3O8HLGzvmKxfvhfwWrb9Xtuv2n4XmAwMLW/A9mW2h9ke1rdv387/FCGEENpVTaKfBmwjaYCkdYCxwKSyYyYBX8puHwrcbdvAFGCIpPWzD4A9WLFvP4QQQp112HVje6mk40hJuxvwK9tzJU0EptueBFwBXCtpHulMfmz22Ncl/Yz0YWFgsu3b6/SzLBcXIEMIoU01ffTYnkzqdinddkrJ7cXAmHYe+2vSEMsQQggNEDNjQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCi4SfQghFFwk+hBCKLhI9CGEUHCR6EMIoeAi0YcQQsFFog8hhIKLRB9CCAUXiT6EEAouEn0IIRRcJPoQQii4SPQhhFBwkehDCKHgqkr0kvaV9KSkeZImVNi/rqQbs/0PSupftn8LSW9L+s/ahB1CCKFaHSZ6Sd2Ai4D9gIHAEZIGlh32VeB121sD5wJnle0/F7hj9cMNIYTQWdWc0Y8A5tl+1vZ7wA3A6LJjRgNXZ7dvBkZJEoCkA4Fngbm1CTmEEEJnVJPoNwMWlNxvybZVPMb2UuBNoLekDYCTgB+sfqghhBA+iGoSvSpsc5XH/AA41/bbq2xAGi9puqTpr7zyShUhhRBCqFb3Ko5pATYvud8PWNjOMS2SugO9gNeAkcChkn4CbAwsk7TY9oWlD7Z9GXAZwLBhw8o/REIIIayGahL9NGAbSQOAF4GxwJFlx0wCvgTcDxwK3G3bwG6tB0g6DXi7PMmHEEKorw4Tve2lko4DpgDdgF/ZnitpIjDd9iTgCuBaSfNIZ/Jj6xl0CCGE6lVzRo/tycDksm2nlNxeDIzp4DlO+wDxhRBCWE0xMzaEEAouEn0IIRRcJPoQQii4SPQhhFBwkehDCKHgItGHEELBRaIPIYSCi0QfQggFF4k+hBAKLhJ9CCEUXCT6EEIouEj0IYRQcJHoQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCi4SfQghFFxViV7SvpKelDRP0oQK+9eVdGO2/0FJ/bPtn5b0sKRHsv/3rm34IYQQOtJhopfUDbgI2A8YCBwhaWDZYV8FXre9NXAucFa2/VVgf9uDgS8B19Yq8BBCCNWp5ox+BDDP9rO23wNuAEaXHTMauDq7fTMwSpJsz7S9MNs+F+ghad1aBB5CCKE61ST6zYAFJfdbsm0Vj7G9FHgT6F12zCHATNv/KG9A0nhJ0yVNf+WVV6qNPYQQQhWqSfSqsM2dOUbSJ0jdOV+v1IDty2wPsz2sb9++VYQUQgihWtUk+hZg85L7/YCF7R0jqTvQC3gtu98P+C/gaNvPrG7AIYQQOqeaRD8N2EbSAEnrAGOBSWXHTCJdbAU4FLjbtiVtDNwOfMf2X2oVdAghhOp1mOizPvfjgCnA48Bvbc+VNFHSAdlhVwC9Jc0DvgW0DsE8Dtga+L6kWdm/f6r5TxFCCKFd3as5yPZkYHLZtlNKbi8GxlR43OnA6asZYwghhNUQM2NDCKHgItGHEELBRaIPIYSCi0QfQggFF4k+hBAKLhJ9CCEUXCT6EEIouEj0IYRQcJHoQwih4CLRhxBCwUWiDyGEgotEH0IIBReJPoQQCi4SfQghFFwk+hBCKLhI9CGEUHCR6EMIoeCqSvSS9pX0pKR5kiZU2L+upBuz/Q9K6l+y7zvZ9icl7VO70EMIIVSjw0QvqRtwEbAfMBA4QtLAssO+Crxue2vgXOCs7LEDSYuJfwLYF7g4e74QQgg5qeaMfgQwz/aztt8DbgBGlx0zGrg6u30zMEqSsu032P6H7eeAednzhRBCyEk1iX4zYEHJ/ZZsW8VjbC8F3gR6V/nYEEIIddS9imNUYZurPKaaxyJpPDA+u/u2pCeriKtdgj7Aq6s86AeVQqutDuNohhiaJY41JIZmiaMZYmiWOJohhhrFsWV7O6pJ9C3A5iX3+wEL2zmmRVJ3oBfwWpWPxfZlwGVVxFIVSdNtD6vV83XlOJohhmaJoxliaJY4miGGZomjGWKodxzVdN1MA7aRNEDSOqSLq5PKjpkEfCm7fShwt21n28dmo3IGANsAD9Um9BBCCNXo8Ize9lJJxwFTgG7Ar2zPlTQRmG57EnAFcK2keaQz+bHZY+dK+i3wGLAU+Dfb79fpZwkhhFBBNV032J4MTC7bdkrJ7cXAmHYeewZwxmrE+EHUrBtoNTVDHM0QAzRHHM0QAzRHHM0QAzRHHM0QA9QxDqUelhBCCEUVJRBCCKHgItGHEELBVdVH38wkfauKw96xfWmd4xhaxWFLbD9Sxxj+1tEhwEu2P16vGLI4muG1+HkVh/3N9sl1jKFZfh8Nj0PSh6s4bJntN+oVQxZHw/OFpIOrOGxxdm20Nm129T56SS8Bl1B5clarL+TwZnqLNBR1VXEMsN2/jjHMtL3D6h5Tgzia4bV4Hjilg8Mm2N62jjE0y++j4XFIWkyaQ7Oqv4lutreoVwxZHA3PF5IWAbd1EMPutreqVZtd/oweuNb2xFUdIGmDHOKYZnvvDuK4u84xHFKjY1ZXM7wW59q+elUHSPpQnWNolt9HM8TxeDUfNnWOAZojX9xh+ysdxPDrWjbY5c/ow8okfYRUU8jAQtsvNzikNVr8PkBSj2wY9modUwSSxti+SdKArNhj/dssQqLP6twfSMmbCbjN9h9zjqMXqRxzaRxT6t3vWNL+9sAvSCUoXsw29wPeAI61PSOPOLJYGv1adCeVzz4I2LQkhtuAK2wvySGGpvh9SHoNuAW4nuWz1nOXVbQdwYp/Ew/lHU+j84WkGbaHtv6fS5tdPdFLOg/4OHANqbYOpDfT0cDTtr+ZUxxHA6cCd7Lim/rTwA9sX5NDDLOAr9t+sGz7TsCltrerdwxZe83wWlxPSqhXs+LfxZeAD9s+PIcYmuX38SRwAXAE0J9USvx62w/k0X4Ww2eAi4GnWfFvYmvSh96dOcXR8Hwh6S5St/n2wP+U77d9QM3bLECif6rShZPs7OEp29vkFMeTwMjyM9asH/jBel8Mztp6ur2fV9K8bGGYumuS1+JJ2//czr6KfzN1iKFZfh9tZ46StiCVKBkLbExaL+K7OcTwOLCf7fll2wcAk+t5UbysvYbni6xm2FDgWuBr5ftt31vrNotwMXaxpBG2y4ulDQfy7O8TFUowA8tY9dX1WrpD0u2ks5XWdQA2J52t5NmN1QyvxeuSxgC/s70MQNJapFIdr+cUQzP9PgCw/QLwE+Ankv6ZrC5VDrqz/Ay61IvA2jnFAE2QL7IFnB6Q9Enbr+TRZhES/TjgEkkbsvwPaXPgb9m+vJwBzJB0J8vf1FuQuit+mEcAtk+QtB9pZa/NSG/wFuCiWo7JrULDXwtSAjuLtHxla2LfGLiHnJJbE/0+7mknvieBH+QUw6+AaZJuYMUPvbGkooh5GUdz5AuAD0k6g9Sd1paLOxqx9kF0+a6bVpI+Ssmbyfb/NiCGDwH7sOKbeortvM4gm0YzvRaSepP+1le9AEuoK0nbsvKH3iTbjzUglmbIF7NJF+sfBtqq+tp+uOZtFSXRh1WTNN5pgZc1nqSPNuKNXRZDU/w+JH3e9h8aHceaSNLDtnfMo61C17qRlNtwwlWR1PA3NPn1ja9Sk7wWeXYVtKcpfh+kvumGknRao2OAhuSL30s6VtImkj7c+q8eDcUZfQ4k7ViPr2NdUbwWoZyk/W3/vtFx5E1SpclStv2xmrdVxEQv6cO2X2t0HM1E0pdtX9noOBpJ0rG2L86xvROA/7K9oMOD6xtH6xKgC23/SdKRwCeBx4HL8pg81oyy60hLbb/V6FjqrZXFRlAAABi4SURBVMsnekm7AL8kDd37CnA6sBVpyNZhtu/PKY4htudkt9cGTiLNAnwUON32u3nEsYr4Xqh3waiStrqRxgf3A/5o+y8l+062fXoOMZRXKRTwHeBHALZ/lkMMbwLvAM+QZqXelNdwurI4fkMa1bE+aRJZT9JM2VGkHPClVTy8VjH0Kb0YLumLLH9/XJ7X7FhJmwJnki4K92T55K1fAWfkNGN6b9t3q50qlrZvqXmbBUj0D5GmuvcEfg8caPs+pVK5F9jeJac4Siel/BToDVxJmmrd2/bROcQwp71dwMdtr1vvGLI4fklKKg8BRwH32v5Wti+Xad9KFTQnA3NZ3h/+78B5ALbrPqwwK9K1I/Ap4HDgANIIi+uBW/I6k5Q0x/aQrCzEi8Cmtt/PJgnNtj0khxhK3x8nA7sB1wGfJ416+Y96x5C1fTcw0fbULNHuBpxMOgn4J9vjc4jhB7ZPlVTpG7bdQcGzD8R2l/4HzCy5/XjZvhkNimMWsHZ2W8CcnGJ4mTStesuyf/1JX9vzei3mlNzuTloL8xZg3dLXqc4xbEGa6n8WsH627dm8XoOsvRll99cmJfvrgVdyjONRYB3gQ8BbpBIQAD3K3zN1jKH0/TED2KDkNXkkx9didtn9h0tuP5Hn30ee/4owYap05NB3yvatk2McvSQdlMWzrrOvgLYtKa+vTX8AetqeVb5D0tScYoCS1932UmC8pFOAu0nfvOrOaQbooZJGA3dJOjePdsusMLIm+5uYBEyStF6OcVwBPAF0A74H3CTpWWAn4IacYlhP0g6k90c32+9Aek0kvb/qh9bUK1m30d2k0szzoa0EQmFHIRah6+YA4E8u6wOXtBVwiO2f5BRH+dewCbZfziZm/Mb2qDziaAZKtbR/7bJqgJK+BlxiO88p70hanzQDdKTt3XNs9+O2n8qrvVXJ+qaxvVDSxqTupBe8cimAerVfPjv3SNsvZZPZptgellMcWwDnAANJ37xPLIljT9u/yyOOvHX5RB+qI6mn7bcbHUdI8v59ZHV+sL0sG4UzCJjvBo9Oy+LqUX6iFmqrsF9V8iZpC0k9stuS9GVJF0g6JrsI1mi5TzNvJWmApIMl/UuObe5bcntjSb+UNEfSdUoLgTRabr8PSQcCLwEvZl1Z/0M6q50jaf+cYlgn6x5pvb+XpG8D+zQ6yUtqlm9dwyRtVpfnjjP62pD0KDDC9ruSziIN8bwV2BvA9biSvnIM7S18LOB7tusy665CHLfaPjC7PZo00mUqaez2j21flUMMpaM8fgn8L3A5cDCwR2t8dY6hWX4fM4H9gPWA2cBw209K2pJU3bPu3SZKdV32tP26pBNJC8JMBvYgXRCdUO8YsjjeIlVWLb1+sj7wLumS2kZ5xFGJpKuBIaRyyTVdLyESfY1Iesz2wOz2w6Q3U2t53NnOYZEJpQWYzwaWVtj9H7Y3rncMWRxtC01L+itpseXnJPUB/jun16I00c+yvX3JvhXu1zGGZvx9PGp7UMm+vIa7trUraTqwm+2/Z992ZziHIZ5Z2xeQVvw60dmSjpKesz0gj/arIWlD13jobTN0KdSFpGOBRaQzlkpvtFpb0DoRgnQlf3Pg+ewiT15mALe6QomB7EJoXkrPHro7WxfT9quSluUUwz9lZ9QCNpIkLz+ryavLsll+H0haKzvx+ErJtm7kNzLtb5IG2X4UeJU0tPPvpByUWxey7eMl7QhcL+lW4EIqr51QV1rFUpu1TvJQ7D56AbuSxm/n4WvA9yX9mfTmmZVNzvgT0N5X+Fr7MvBCO/tyGdWQ2U7S37KvydtnI49ap+J3yymGy4ENScM5rwb6ZDF8lDTaIg/N8vsYT5bQy0bZbE6aJZqHbwC/kXQN8H/AdEm/Au4jm62cl+yD91PZ3XtJHzq5UVpqcwawJ6nbaANgL+DhbF/t2+zqXTeSvmn7fEm7uGSqfQPj2Za0JmXrijrTWrtw1nTZsL5tnVNZitBcsm8Qn2HF90duC8a3E9MmwA7OcSEYNWCpzSIk+lm2t8+rr7GZZWerp5Lq/pwCHE+aFPI48E3bL+UYS/fWLjNJPYF/Ic1MbXixOeVU4E3SLaRvlLc2cmirpI1Ikwn7AXfYvq5k38W2j21UbI2mtGbtDsBjtp/Iqc2nSNfw3izb3guY7jqsW1uErpvHJc0H/jkbPtf67xG1X/ulqK4iDdtbQFo+7u/A50jD6X6RVxCSxgEvS3pKaSm9OaRSBLMlHZFXHKuQ1/J5I0m1jl6Q9FtJB2XdV3m7ktSV+TtgrKTfSWqte7RTHgGUDXftJemKRgx3zfrlW2+PJs2Q3R+4Lfu7zUPrUpuXSPpu9u8XpO6cM+rRYJc/o4e2M9kppDoiK7D9fP4RNUbZ6IoVqlXmNdIka+sRUp/jhqThfDvYfiZ7Q9+VxwiLVXzI51bgrfX3obQ+6YHAEaSFPv4AXG/7znrHkMVRPuroe8BnSe+Xu3IaddPw4a5Z2w0fEZa1netSm4UYdeO0LFwuv6AmV/oN7ZpV7Ku3951K0r4q6W3bzwBkJSHyiuEjpDdS+RtHwF9zisEA2SiKa4FrlVYQOgyYAOSS6IF1S0bdYPsMSS3An8mp9lCZYSUfPOdKqnuZ5BLNMCKMLKHnVWeoGIkeQKku/Wmkao3dSW9ouw6rtXQyrqtJkzEuyoaW1dNtyqbW2z65JIatgTxn/70g6cekM/onlMo230Ia6ZDXdYJmKPC2Ur98do3iF+TYlUYq3703aQRYaxxXS3oZuCCnGJphuCtkI8KyONZVtn5wziPC2iXpMtehVHIhum4AJD0B/Acrr6i+qGFBAZKGk0rmjrB9UiNjyUt28e/fSGdPF5LOrL8MPE9ahCW3i8KhOUg6tWzTxbZfybpdf+Ic1mtYlWYZEaY6LbVZpET/oO2RjY4jhFKSRpC+WU6TNJA0SeaJPIfzVYhpV7LVnfK6ThAaqwijblrdI+lsSTtLGtr6L6/Gs5EEZ0p6QtKi7N/j2bZcprqH5SQNkfSApAWSLssufrXuy6s076nAz4FLsq6sC0l94hOyC6K5KP15Jf1rFseGwKmS8qoxc4Kkfnm09UFlgwjyaGctSV+RdLuk2ZIelnSDpD3r1maBzujL611DOpPaO6f2p5CGal2dXRxuHQ00Dhhl+9N5xBESSfeR1g9+gDRr+cvAAdnon7aRF3WO4RHSil/rkkaZ9LP9N6VFRx7Msb5L6UiTacBns26TDYAHbA/OIYZmWT+34jqtpD77X9jum0MMV5K6Mf8EHAr8jTQE+iTgNts1v25SmETfaJKetP3Pnd2XB0nDgJdsv9jhwQVRYUjhXqQlDY8i9Q/nMaSwNMGu8OGS83DX2aTp9mtRtshHjh96zbJ+7hLgN1Sub3Oo7Q1ziGFO6Ye8pAds75TNbZhle9tat9nlR91I+qLtX6udkrC2f5ZTKM9L+n+kM/rWqngfIZ3RL8gphvYcDwyRVPPyp52hfAvNSVKv1tmHtu+RdAhp0lAu5YGB9ySt71RvfceSwHqRZi/npRcpqQpwyUiTntm2PDgb3nkncKektUmlk48g1cav+5l0Zg5wTqURcJI+VeH4elgiaavs2+VQ4D0A2/9QnZYd7fKJnlQQCFKfYyMdThobfa+kf8q2vUxaI/SwhkUF2P4SpPKnjYyD5YXmvkCFyW01dhawLanrBgDbcySNAr5f57Zb7W77H1nbpYl9bSC3seO2+7ezaxmpLnwemmX93H8ndZVUktdrcSLpmuJi0t/CWABJfUnDgmsuum4KRqsof9rQwMIaTU20fm4zUJo52DubWFh3RRp100bSjEbHUCqv0T9qQPnTVcSyj1Itj0mSbstu79vxI0MRdYUkL+mUvNpy0pbklco3100hz+jzusBULUmX2/7XHNrJvfxpO3GcRypFew2phgekyolHA0/b/mYecYTQGSqrD1XHdiaVbyKdkN0NYLvm3ZpF6KOv5PZGB1AqjySfEZVHEywjv4tukIbvrfShIulGUimGSPShIbLyBxV3kdbUzUM/UpXZX7J8/dphwE/r1WCXT/TSCjUzACit89LeMXWK5aNZ+/+bXVjZDXjS9tx6t51pLX96J8tH+mwBfBr4YU4xACyWNMIrrmYEqXLj4hzjWEnOI3/ai+FPwBJS/aO6XHzrSnHk7A1SLfiXy3dIymt03DDSyc73SGvXzpL0d9v31qvBLp/oSVevf0eaaNC2bJtSkaJdSaMb7iHVaq8bSV8njbqRpLNIwyrnAj+W9BPbV9SzfWgrVDWJFcufTgW+U6/yp+0YR5oNuiHLu242J412GJdjHJXkOfKnPUcDm5BTLfhmjqMBHzbXkAofrpTogesqbKu5bATWuZJuyv5/mTrn4i7fRy+pB2nB4y8AA0if2OuRLjTfSfoDqvsaodksyJFZ288DW2dn9h8C7slrckwzyb7htNXbbp0xvCZSKk/snD9wmzaOVpI2JfuwsX1Ro+NpBEmfBz5p+7t1a6OrJ/pS2SSMPsDf8x5OqBUXVpjtkgUMmuHisOpU/rSZSdqHtOBH6VDT22z/Maf2twB+AowinYAI2Ih00W2C7flrUhyhfZI+7Dous1mo4ZW2l9h+qUFjxpdlHzSQlu8D2r5xNMPrfGmjA4D8hr5mI3++CdxLSnJnZ7dPkHR+HjEANwL/BXzU9ja2tyadvd5KjotONEMckmZIOlnSVnm018wkla4VMVBpDdnpkuZLqksF3kKd0TdSdta0sPwCn6TNSHWu/1T5kaEesnIPlUb+CHjKdViAuUJbT7fXzqr2FTEOSc+Ryk8cRirwdj1wo+2F9W672ZR9+78duND2HUolrc+z/clat9kMZ5pFsaDSKA7bL7Ym+SzJ1I2k0kJJa2dnUJMk/UjS+vVsuwktzt445fIc+fOwpIsljZS0afZvpKSLgZk5xdAscbxu+z+zcerfBrYhjRC7R9Ia1aVYZlPbdwBko9TqMsQzzuhrRGl5ug5H/9i+qo4xlJ4p/BToDVxJ6qfu7Qav4gPporXzKYs7FLiEVAOpfOTPsa7DKj4VYlgH+CowmuUXpReQlva7orUOzpoQR+nfZsm2bqShv4fb/nK9Y1gVSY9nNy+yfWGd23qDtF6vSCOetswK3yHpUduDat5mJPraaGf0Tw/SOpS5jP7RimVxZ5HGCy/JvknMdn71zxte87sklhj50wQk3WB7bKPjWBVJvUmjf+o64VLSHmWbHrb9tlK120PrMfooEn0dNGr0j6RnSV+L1yKtzbptyb4VRgLVOY6G1/xudpI+3wyTlJoljrxlSbVtNFalCVRFUoQJU00nK8HaiAWw72X5JKAHJH3E9svZWW0uVfIyzVDzu12VuhEaYDh1KknbSQ2PQ9JQ23mNxtoe+AWpRn/rQjz9su6UY/OKoz31GgYdZ/Sh5iTtBjxfeq2iZN8w29MbEFZoUsqp6F/W1izg67YfLNu+E3BpHt96s0lrFXeRulhrvrZuJPoCyYZ4/p/txVm//DhgKKmA0uWNqu0SlpN0Td4XxSUdANxpu6F1hppBB0NN52VzDOodw/uk2fOlo/Bai5ttZnudWrcZXTfFMhloHVJ4JrAVaVLM3qSv6F9pUFxtJJ1ie2IO7WxOmiS1GXAHcHbWpYakW20fmEMMFcvRStoY6lOOth03Au9IuoM0fn2K7fdzaruNpN2Bl20/KWlX0oiTx+t98bPMHdnY9WtYXvhvc1Ldn1xmTAPPAqPa+cZbl8JqcUZfIJIesz0wu/0wadTNsux+bhdjV0X51fy+izTc9QHS0MIdgf1tL8qrJEU2C7i8HO31ZEvH1bNaYVkcM0kf9odmbQ8izZS9PscYziOdhHQHppDKMdwB7AHMtH1iHnFksezHikNNW4BJtifn1P6/AffZnl1h3/G2L6h5m5Hoi0PSFOAs23crVfT8lu3ns2Fjd+c46maVNb9t1/2bpKRZpYXkJH0R+A7pYvVNeVyMlbQWqQzDZ1lejvZZ2x+rd9tlcaxw8Tm7OH8YaWHufrY3zyGGuaQPmPVIF0E3s/1uNkJtZj3GjoflouumWL4GXCPpNOBNYFZ2Nvch4Fs5xtEMNb/XltSjtV/a9q8l/S/pbHKDVT+0NtyAcrTtKF+Y+3+BnwM/l7RlTjHYtiW1LpLeeoa5jCaZoS9pvO3LGhzDp23fVevnjURfILYXkPqAtyUt5XcV6WvptNYunJw0vOY3qbtkJGnIKQC2/yRpDKnIWW5stwBjlMrRtvdtp57+o70dtp/PKYbbJf0PaRLhL4HfSnqA1HXz55xi6Eieq7C15wrSYkE1FV03BSJ1vJJWNceE+lGdy9E2M0k7k87sH1CqYnkQ8AJwc84nIg1V4SJ92y5gb9s1/8YZib5AmqHeTkmbDV1WUVIf26+W3P8i6WLgo6ShpnksLXmy7dOz2wNJI6C6k7oqDi8fy90IedUeaqfthn/oZaN/RgCP2r4zpzZfB74IvF2+i1TR8yM1bzMSfXE0Q72dLI62ZRWB0mUVdwFyWVaxrMDbyaQPmuuAz5Nq3rTbnVGnGHIpR9tOHA2vPSRpF1KXzTLS3+jppOG/awOH2b6/3jFkcTxke0R2+1+BfyONQPoM8HvbZ+YQwx2k98E9Ffb92fbuNW8zEn0xNareTtZ2w5dVLCvwNgPYzfY72esyI4+z2LJEv8KQzryGeGZtNbz2kKSHSMNce5KqZh5o+z6lKqMX2N6l3jFkcZT+XUwDPmv7FUkbAA806ttNvcXF2IJqYL0dgCVOZVfflfRMNsoD269LyuvMYj1JO5C6SbrZfieLYUk2MzEPH8v6Y0Wqp7J+9rpAOpPNSzPUHlrb9iNZm6/Yvg/A9gxJdanB3o61shOOtUgnuq9kcbwjKZeZ4424lhaJPtTDMklrZx82jVpW8SXgZ9nt1yRtYvulbE5BXqUgRpfdXwvaKideklMMAP9O+6N9DsophtLf+3fK9tV8yv8q9AIeJn34WtJHs2+bPclv1M092TyXVV5LI42aq4nougk1pyZeVjGbxNSj5Mw65ECp3s6fyl/3bPTNIbZzHfJaTmkFto/Yfi6HtnK/lhaJPtRcswzzVHPUVqlIdSpH205bPwN+Z/svebTXlUjaGtiO9HfxWAPaz+VaWiT6UHPNMMyzGWqrqAHlaNuJ4xXSRfG+pAJn19vOc81alJYN/BrQD/hj6YdO6TDUHOK4Bxhj+1VJRwHfJ03YGglc5jrUmWkGkehDzTXDMM9mqK2iBpSjbSeOmbZ3kLQNqajZWNLv4npS0n8qhxh+CawPPAQcBdxr+1vZvtwWglHJmqzZqJt9nQrdrU8adZPLcpt5i4uxoeac6stcDFzcwGGethteWyX3crTtMIDtp4EfAj+UNIRU1GwyUPca7MCI1iQq6ULS38YtWQx5lh5YImkz2y+SJiy9k23/B+nDr5Ai0Ye6auAwz2aorXIeqaDcSomefOvtrJRIbc8hDbssHwFTL23fXrKL9OMlnQLcTRpbn5f/AO7MRr3MBe6W9EfShLorc4wjV9F1EworaqskknraLp9un3cMvwZ+bfuPZdu/BlxiO7d5BZJ6AUeSCv91JxX+u832E3nFkLdI9KGQmmjkz0ZAX9vPlG0fkp1V56LRtYdCYzVFHegQ6uAeScdnY/rbSFpH0t6SriaN/qkbSYcBTwC/kzRX0vCS3VfVs+2yOL4O3A88IOkY4A+kmj+3SPpqXnGUxTRA0sGS/iXndoeU3F5b0smSJkn6UXZBtpDijD4UUjsjf9YjndzkNfJnFrBfNiN3BKlO/3dt35JzrZtmqD3Utk6vpNGk6xdTgU8CP67nUNuyOErrD/0U6E3qmz8Q6O2cF27PS1yMDYXUJCN/utl+KYvnIUl7AX+Q1I/KBcbqpRlqD5WuZHUSqe76c5L6AP9Nft9wSi9MjyKthLZE0p+BldZwLYpI9KHwGjjy5y1JW7X2z2dn9nuS6tJ/Isc4mqH2UOkHSvfWUgPZxKU8L4z3knQQ6edeN3tNyIbiFrZ7IxJ9CPVzDGWJ1PZbkvYlLc6dl4NZPpa+pWR7b+DbOcWwndKi8QLWLSkmtg75jl+/l7RAPKRrFh+x/XJ2sfrVVTyuS4s++hDqpIlG/jRFHO20uzGp0F0uC4+sqWLUTQj10/CRP80Sh6SKs19tv9Ga5Ns7psZxHCBp3Xq302zijD6EOmmGmj/NEkczFLrL2vs7qezBHaRaP1Ns57UQTcNEog8hBw0c+dMUcTTDh00Wx0xgb+BQUnG3QaQ1Y6+3fW+922+USPQhhFw18kOvvFJmdhH2MFJxtX62N88znrxEog8hrDFWNVFN0pa2n887pjxEog8hrDEk7Wl7aqPjyFsk+hDCGqOZh5rWUwyvDCGsSRo+1LQR4ow+hLDGaJbRP3mLRB9CWCM1y5DXPESiDyGEgos++hBCKLhI9CGEUHCR6EMIoeAi0YcQQsFFog8hhIL7/yEapYuE5R01AAAAAElFTkSuQmCC\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"time_from_last_sett_pl_hour PSI: 0.1198387217423427\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>time_from_last_sett_pl_hour_base</th>\n",
" <th>time_from_last_sett_pl_hour_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 0.0442]</th>\n",
" <td>0.061662</td>\n",
" <td>0.030916</td>\n",
" <td>0.030746</td>\n",
" <td>0.690389</td>\n",
" <td>0.021227</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0442, 0.109]</th>\n",
" <td>0.061242</td>\n",
" <td>0.045800</td>\n",
" <td>0.015442</td>\n",
" <td>0.290547</td>\n",
" <td>0.004487</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.109, 0.394]</th>\n",
" <td>0.061314</td>\n",
" <td>0.066389</td>\n",
" <td>-0.005075</td>\n",
" <td>-0.079522</td>\n",
" <td>0.000404</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.394, 3.631]</th>\n",
" <td>0.061392</td>\n",
" <td>0.102944</td>\n",
" <td>-0.041552</td>\n",
" <td>-0.516905</td>\n",
" <td>0.021478</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.631, 20.221]</th>\n",
" <td>0.061404</td>\n",
" <td>0.095043</td>\n",
" <td>-0.033639</td>\n",
" <td>-0.436853</td>\n",
" <td>0.014695</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(20.221, 65.875]</th>\n",
" <td>0.061398</td>\n",
" <td>0.081077</td>\n",
" <td>-0.019679</td>\n",
" <td>-0.278017</td>\n",
" <td>0.005471</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(65.875, 187.343]</th>\n",
" <td>0.061404</td>\n",
" <td>0.040817</td>\n",
" <td>0.020587</td>\n",
" <td>0.408377</td>\n",
" <td>0.008407</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(187.343, 526.391]</th>\n",
" <td>0.061398</td>\n",
" <td>0.071340</td>\n",
" <td>-0.009942</td>\n",
" <td>-0.150073</td>\n",
" <td>0.001492</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(526.391, 1446.336]</th>\n",
" <td>0.061404</td>\n",
" <td>0.105075</td>\n",
" <td>-0.043671</td>\n",
" <td>-0.537200</td>\n",
" <td>0.023460</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1446.336, inf]</th>\n",
" <td>0.061404</td>\n",
" <td>0.049112</td>\n",
" <td>0.012293</td>\n",
" <td>0.223382</td>\n",
" <td>0.002746</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.385978</td>\n",
" <td>0.311488</td>\n",
" <td>0.074490</td>\n",
" <td>0.214419</td>\n",
" <td>0.015972</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" time_from_last_sett_pl_hour_base \\\n",
"(-inf, 0.0442] 0.061662 \n",
"(0.0442, 0.109] 0.061242 \n",
"(0.109, 0.394] 0.061314 \n",
"(0.394, 3.631] 0.061392 \n",
"(3.631, 20.221] 0.061404 \n",
"(20.221, 65.875] 0.061398 \n",
"(65.875, 187.343] 0.061404 \n",
"(187.343, 526.391] 0.061398 \n",
"(526.391, 1446.336] 0.061404 \n",
"(1446.336, inf] 0.061404 \n",
"NaN 0.385978 \n",
"\n",
" time_from_last_sett_pl_hour_prod base-prod \\\n",
"(-inf, 0.0442] 0.030916 0.030746 \n",
"(0.0442, 0.109] 0.045800 0.015442 \n",
"(0.109, 0.394] 0.066389 -0.005075 \n",
"(0.394, 3.631] 0.102944 -0.041552 \n",
"(3.631, 20.221] 0.095043 -0.033639 \n",
"(20.221, 65.875] 0.081077 -0.019679 \n",
"(65.875, 187.343] 0.040817 0.020587 \n",
"(187.343, 526.391] 0.071340 -0.009942 \n",
"(526.391, 1446.336] 0.105075 -0.043671 \n",
"(1446.336, inf] 0.049112 0.012293 \n",
"NaN 0.311488 0.074490 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, 0.0442] 0.690389 0.021227 \n",
"(0.0442, 0.109] 0.290547 0.004487 \n",
"(0.109, 0.394] -0.079522 0.000404 \n",
"(0.394, 3.631] -0.516905 0.021478 \n",
"(3.631, 20.221] -0.436853 0.014695 \n",
"(20.221, 65.875] -0.278017 0.005471 \n",
"(65.875, 187.343] 0.408377 0.008407 \n",
"(187.343, 526.391] -0.150073 0.001492 \n",
"(526.391, 1446.336] -0.537200 0.023460 \n",
"(1446.336, inf] 0.223382 0.002746 \n",
"NaN 0.214419 0.015972 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"trx_denied_co_90 PSI: 0.967668393694691\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>trx_denied_co_90_base</th>\n",
" <th>trx_denied_co_90_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 1.0]</th>\n",
" <td>0.659671</td>\n",
" <td>0.282932</td>\n",
" <td>0.376739</td>\n",
" <td>0.846533</td>\n",
" <td>0.318922</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1.0, 2.0]</th>\n",
" <td>0.100508</td>\n",
" <td>0.071045</td>\n",
" <td>0.029464</td>\n",
" <td>0.346933</td>\n",
" <td>0.010222</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(2.0, 3.0]</th>\n",
" <td>0.065216</td>\n",
" <td>0.061537</td>\n",
" <td>0.003679</td>\n",
" <td>0.058060</td>\n",
" <td>0.000214</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(3.0, 6.0]</th>\n",
" <td>0.094659</td>\n",
" <td>0.142679</td>\n",
" <td>-0.048020</td>\n",
" <td>-0.410317</td>\n",
" <td>0.019703</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(6.0, inf]</th>\n",
" <td>0.079946</td>\n",
" <td>0.441807</td>\n",
" <td>-0.361861</td>\n",
" <td>-1.709518</td>\n",
" <td>0.618608</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" trx_denied_co_90_base trx_denied_co_90_prod base-prod \\\n",
"(-inf, 1.0] 0.659671 0.282932 0.376739 \n",
"(1.0, 2.0] 0.100508 0.071045 0.029464 \n",
"(2.0, 3.0] 0.065216 0.061537 0.003679 \n",
"(3.0, 6.0] 0.094659 0.142679 -0.048020 \n",
"(6.0, inf] 0.079946 0.441807 -0.361861 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, 1.0] 0.846533 0.318922 \n",
"(1.0, 2.0] 0.346933 0.010222 \n",
"(2.0, 3.0] 0.058060 0.000214 \n",
"(3.0, 6.0] -0.410317 0.019703 \n",
"(6.0, inf] -1.709518 0.618608 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEdCAYAAAAcmJzBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAfCUlEQVR4nO3dfXRU9b3v8fenQaRW8ViJPUqg0FZtUSLBELRwUFGLXlxYPVpBrVVq8dbiQ7kP0FOLaG+71Nrn0nvlVFqxrdjrQ6WKSjmAtGdVAZWLPEgFRYlSpQi0VJEHv/ePGZIhTJKBTGYnez6vtbLI7P2bPd/8VvjkN7/Z+7cVEZiZWef3gaQLMDOz4nCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSnRJ6oV79OgRffr0Serlzcw6pWefffavEVGZb19igd6nTx+WLFmS1MubmXVKkl5tbp+nXMzMUsKBbmaWEg50M7OUSGwO3SwNdu7cSX19Pdu3b0+6FEuZbt26UVVVxUEHHVTwcxzoZm1QX1/PYYcdRp8+fZCUdDmWEhHBpk2bqK+vp2/fvgU/z1MuZm2wfft2jjzySIe5FZUkjjzyyP1+5+dAN2sjh7m1hwP5vXKgm5mlRKeeQ+8z6bE2H2PdbSOLUIlZRjF+J3O19vu5ZcsWfv3rX3PttdcW9XWvvPJKzjvvPC666KL9fu6SJUuYMWMGP/rRjwp+zumnn86dd95JbW3tfr9eITZv3szYsWNZu3Yt3bp1Y/r06Zx44okAPPHEE9xwww3s3r2bq6++mkmTJjV7nD0XRPbo0aNd6mwrj9DNOrEtW7bw05/+dJ/tu3fvTqCajNra2v0K81L49re/zYABA1i2bBkzZszghhtuADL99JWvfIXHH3+clStXct9997Fy5cqEqz1wDnSzTmzSpEmsXbuWAQMGMGjQIM444wwuvfRS+vfvz+LFi6murmb79u384x//4IQTTmD58uV5jxMRjB8/nn79+jFy5Ejeeuuthn3PPvssp512GieffDIjRoxgw4YNQGZUPXHiROrq6jjuuOP4wx/+AMCCBQs477zzAPjHP/7B2LFjGTRoEDU1NTzyyCMAvPvuu4wePZrq6mouueQS3n333RZ/zieeeIKBAwdy0kknceaZZwLw9ttv89nPfpbq6mpOOeUUli1b1uzzV65c2fC8T37yk6xbt44333yTRYsW8YlPfIKPfexjdO3aldGjRzfU2JzvfOc71NXVUVdXx5o1awD43e9+x+DBg6mpqeGss87izTffBOCpp55iwIABDBgwgJqaGv7+9783HGPQoEFUV1dz8803t/h6+6NTT7mYlbvbbruN5cuXs3TpUhYsWMDIkSNZvnx5w6luo0aN4qabbuLdd9/l8ssvb5hmaOrhhx9m9erVvPDCC7z55pv069ePsWPHsnPnTq677joeeeQRKisruf/++/n617/O9OnTAdi1axeLFi1i9uzZ3HLLLcydO3ev437rW99i+PDhTJ8+nS1btlBXV8dZZ53FXXfdxSGHHMKyZctYtmwZAwcObPZn3LhxI1/60pdYuHAhffv25e233wbg5ptvpqamht/+9rfMmzePK664gqVLl+Y9xkknncRDDz3E0KFDWbRoEa+++ir19fW8/vrr9OrVq6FdVVUVzzzzTIt93r17dxYtWsSMGTO48cYbefTRRxk6dChPP/00kvjZz37GHXfcwXe/+13uvPNOpk6dypAhQ9i2bRvdunVjzpw5vPTSSyxatIiIYNSoUSxcuJBhw4a1+LqFcKCbpUhdXd1e5y1PnjyZQYMG0a1btxanQRYuXMiYMWOoqKjgmGOOYfjw4QCsXr2a5cuXc/bZZwOZKYqjjz664XkXXnghACeffDLr1q3b57hz5sxh1qxZ3HnnnUDmNM/XXnuNhQsXcv311wNQXV1NdXV1s7U9/fTTDBs2rOHn+vCHPwzAH//4Rx588EEAhg8fzqZNm9i6dSuHH374PseYNGkSN9xwAwMGDKB///7U1NTQpUsXImKftq2dXTJmzJiGf7/61a8CmesRLrnkEjZs2MCOHTsaah0yZAgTJkzgsssu48ILL6Sqqoo5c+YwZ84campqANi2bRsvvfSSA93M9vahD31or8dvv/0227ZtY+fOnWzfvn2f/bnyBVlEcMIJJ/CnP/0p73MOPvhgACoqKti1a1fe5z/44IMcf/zxBb1ePhHRbG2FHrN79+78/Oc/b3he37596du3L++88w7r169vaFdfX88xxxzTYj25r7Hn++uuu44JEyYwatQoFixYwJQpU4DMH5KRI0cye/ZsTjnlFObOnUtE8LWvfY1rrrmm5R/8AHgO3awTO+ywwxrmZfMZN24c3/zmN7nsssuYOHFis+2GDRvGzJkz2b17Nxs2bGD+/PkAHH/88WzcuLEh0Hfu3MmKFSsKrm/EiBH8+Mc/bgjf559/vuH1fvWrXwGwfPnyFue/Tz31VJ566ileeeUVgIYpl9xjLFiwgB49etC9e/e8x9iyZQs7duwA4Gc/+xnDhg2je/fuDBo0iJdeeolXXnmFHTt2MHPmTEaNGtXiz3T//fc3/HvqqacCsHXrVnr27AnAPffc09B27dq19O/fn4kTJ1JbW8uLL77IiBEjmD59Otu2bQPg9ddf3+szi7bwCN2siEp9GuyRRx7JkCFDOPHEE/ngBz/IRz7ykYZ9M2bMoEuXLlx66aXs3r2bT3/608ybN69hOiXXBRdcwLx58+jfvz/HHXccp512GgBdu3blgQce4Prrr2fr1q3s2rWLG2+8kRNOOKGg+r7xjW9w4403Ul1dTUTQp08fHn30Ub785S9z1VVXUV1dzYABA6irq2v2GJWVlUybNo0LL7yQ999/n6OOOorf//73TJkypeEYhxxyyF5B2tSqVau44oorqKiooF+/ftx9990AdOnShZ/85CeMGDGC3bt3M3bs2FZ/tvfee4/Bgwfz/vvvc9999wEwZcoULr74Ynr27Mkpp5zS8MfnBz/4AfPnz2943XPPPZeDDz6YVatWNfwxOPTQQ/nlL3/JUUcdVVCftkT53raUQm1tbbT1Bhc+D92StmrVKj71qU8lXYalVL7fL0nPRkTeE/Y95WJmlhKecjErIy+88AKf//zn99p28MEHt3qqXqkMHjyY9957b69t9957L/379y/o+T//+c/54Q9/uNe2IUOGMHXq1P2q44ILLmiYNtnj9ttvZ8SIEft1nFJzoJuVkf79+zd7rnZH0NY/LFdddRVXXXVVm+t4+OGH23yMJHjKxcwsJRzoZmYp4UA3M0uJggJd0jmSVktaIynv2pKSPidppaQVkn5d3DLNzKw1rX4oKqkCmAqcDdQDiyXNioiVOW2OBb4GDImIzZLafoa8WWc0Zd91RNp2vK0t7vZ66IUp1nroxVbs9dULGaHXAWsi4uWI2AHMBM5v0uZLwNSI2AwQEcW5jtXMWuT10AtTyvXQ861pUyqFBHpPYH3O4/rstlzHAcdJ+k9JT0s6p1gFmlnzvB56addD79OnT8PPnLse+pVXXsmECRM444wzmDhxYrO1bdq0ic985jPU1NRwzTXX5F1grC0KOQ893/JlTavoAhwLnA5UAX+QdGJEbNnrQNI4YBxA796997tYM9ub10PvGOuhA/z5z39m7ty5VFRUcN111+Wt7ZZbbmHo0KFMnjyZxx57jGnTprX4WvurkECvB3rlPK4C3sjT5umI2Am8Imk1mYBfnNsoIqYB0yCzlsuBFm1m+Xk99GTWQwe4+OKLqaioaLG2hQsX8tBDDwEwcuRIjjjiiBZfa38VEuiLgWMl9QVeB0YDlzZp81tgDPALST3ITMG8XMxCzax1Xg89mfXQYe++b6m2Qn/uA9HqHHpE7ALGA08Cq4DfRMQKSbdK2rNw8JPAJkkrgfnA/4iITe1VtJlleD30jrEeelPN1Za7/fHHH2fz5s0tvtb+Kmgtl4iYDcxusm1yzvcBTMh+mZWvVk4zLDavh94x1kNvqrnabr75ZsaMGcPAgQM57bTTiv5ZotdD93ro1gZeD728FPu88dZ4PXQzszLl5XPNyojXQy9Mc+uh5zuTpyNxoJuVEa+HXhivh25WppL6HMrS7UB+rxzoZm3QrVs3Nm3a5FC3oooINm3aRLdu3fbreZ5yMWuDqqoq6uvr2bhxY9KlWMp069aNqqqq/XqOA92sDQ466KC9LrU3S5KnXMzMUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczS4mCAl3SOZJWS1ojaVKe/VdK2ihpafbr6uKXamZmLWn1BheSKoCpwNlAPbBY0qyIWNmk6f0RMb4dajQzswIUMkKvA9ZExMsRsQOYCZzfvmWZmdn+KiTQewLrcx7XZ7c19a+Slkl6QFKvfAeSNE7SEklLfA9GM7PiKiTQlWdb01uc/w7oExHVwFzgnnwHiohpEVEbEbWVlZX7V6mZmbWokECvB3JH3FXAG7kNImJTRLyXffjvwMnFKc/MzApVSKAvBo6V1FdSV2A0MCu3gaSjcx6OAlYVr0QzMytEq2e5RMQuSeOBJ4EKYHpErJB0K7AkImYB10saBewC3gaubMeazcwsj1YDHSAiZgOzm2ybnPP914CvFbc0MzPbH75S1MwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFKioNUWzczK3pTDi3CMrW0/Rgs8QjczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYpUVCgSzpH0mpJayRNaqHdRZJCUm3xSjQzs0K0GuiSKoCpwLlAP2CMpH552h0GXA88U+wizcysdYWM0OuANRHxckTsAGYC5+dp903gDmB7EeszM7MCFRLoPYH1OY/rs9saSKoBekXEo0WszczM9kMhga4826Jhp/QB4PvAf2v1QNI4SUskLdm4cWPhVZqZWasKCfR6oFfO4yrgjZzHhwEnAgskrQNOAWbl+2A0IqZFRG1E1FZWVh541WZmto9CAn0xcKykvpK6AqOBWXt2RsTWiOgREX0iog/wNDAqIpa0S8VmZpZXq4EeEbuA8cCTwCrgNxGxQtKtkka1d4FmZlaYgm5wERGzgdlNtk1upu3pbS/LzMz2l68UNTNLCQe6mVlKONDNzFLCgW5mlhIOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaVEQYEu6RxJqyWtkTQpz/7/KukFSUsl/VFSv+KXamZmLWk10CVVAFOBc4F+wJg8gf3riOgfEQOAO4DvFb1SMzNrUSEj9DpgTUS8HBE7gJnA+bkNIuJvOQ8/BETxSjQzs0J0KaBNT2B9zuN6YHDTRpK+AkwAugLDi1KdmZkVrJARuvJs22cEHhFTI+LjwETgprwHksZJWiJpycaNG/evUjMza1EhgV4P9Mp5XAW80UL7mcBn8+2IiGkRURsRtZWVlYVXaWZmrSok0BcDx0rqK6krMBqYldtA0rE5D0cCLxWvRDMzK0Src+gRsUvSeOBJoAKYHhErJN0KLImIWcB4SWcBO4HNwBfas2gzM9tXIR+KEhGzgdlNtk3O+f6GItdlZmb7yVeKmpmlREEj9FSbcngbn7+1OHWYmbWRR+hmZinhQDczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIFBbqkcyStlrRG0qQ8+ydIWilpmaT/kPTR4pdqZmYtaTXQJVUAU4FzgX7AGEn9mjR7HqiNiGrgAeCOYhdqZmYtK2SEXgesiYiXI2IHMBM4P7dBRMyPiHeyD58GqopbppmZtaaQQO8JrM95XJ/d1pwvAo+3pSgzM9t/XQpoozzbIm9D6XKgFjitmf3jgHEAvXv3LrBEMzMrRCEj9HqgV87jKuCNpo0knQV8HRgVEe/lO1BETIuI2oioraysPJB6zcysGYUE+mLgWEl9JXUFRgOzchtIqgHuIhPmbxW/TDMza02rgR4Ru4DxwJPAKuA3EbFC0q2SRmWbfQc4FPi/kpZKmtXM4czMrJ0UModORMwGZjfZNjnn+7OKXJeZme0nXylqZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKVHQaotmVqamHF6EY2xt+zGsIB6hm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5QoKNAlnSNptaQ1kibl2T9M0nOSdkm6qPhlmplZa1oNdEkVwFTgXKAfMEZSvybNXgOuBH5d7ALNzKwwhVz6XwesiYiXASTNBM4HVu5pEBHrsvveb4cazcysAIVMufQE1uc8rs9uMzOzDqSQEbrybIsDeTFJ44BxAL179z6QQ1gz+kx6rM3HWHfbyCJUYmZJKWSEXg/0ynlcBbxxIC8WEdMiojYiaisrKw/kEGZm1oxCRuiLgWMl9QVeB0YDl7ZrVWZWFG1957auW5EKsZJodYQeEbuA8cCTwCrgNxGxQtKtkkYBSBokqR64GLhL0or2LNrMzPZV0A0uImI2MLvJtsk53y8mMxVjZmYJ8ZWiZmYp4VvQmVlZKIfPEzxCNzNLCQe6mVlKeMrFGrX1Du++u7tZojxCNzNLCY/QLXXa/OGXl0CwTsqBbtZUW6eewNNPlghPuZiZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIFBbqkcyStlrRG0qQ8+w+WdH92/zOS+hS7UDMza1mrgS6pApgKnAv0A8ZI6tek2ReBzRHxCeD7wO3FLtTMzFpWyAi9DlgTES9HxA5gJnB+kzbnA/dkv38AOFOSilemmZm1RhHRcgPpIuCciLg6+/jzwOCIGJ/TZnm2TX328dpsm782OdY4YFz24fHA6mL9IG3QA/hrq63Kg/siw/3QyH3RqKP0xUcjojLfjkLuKZpvpN30r0AhbYiIacC0Al6zZCQtiYjapOvoCNwXGe6HRu6LRp2hLwqZcqkHeuU8rgLeaK6NpC7A4cDbxSjQzMwKU0igLwaOldRXUldgNDCrSZtZwBey318EzIvW5nLMzKyoWp1yiYhdksYDTwIVwPSIWCHpVmBJRMwC7gbulbSGzMh8dHsWXWQdagooYe6LDPdDI/dFow7fF61+KGpmZp2DrxQ1M0sJB7qZWUo40M3MUqKQ89BTQdKPCmj2t4i4qd2LSZikv7XWBNgQEceVop4kSVpWQLONEXFmuxeTMPdFo876f6RsAp3M8gSTW2kzCUh9oANrI6KmpQaSni9VMQmrAP5LC/vFvqfpppX7olGn/D9SToH+/Yi4p6UGko4oVTEJ+9citUmDayLi1ZYaSLq2VMUkzH3R6NIC2nS4/yM+bbGMSfoI0JPMMg1vRMSbCZeUKEkfBiIiNiddS9LKvS8kPRcRAyXdGxGfT7qeQpXNCD27JMEXgQuAY8iGGPAIcHdE7EywvJKSNAD4P2SWaHg9u7lK0hbg2oh4LrHiSkxSb+AO4ExgS2aTugPzgEkRsS7B8krKfbGXrpK+AHxa0oVNd0bEQwnU1KqyGaFLuo/ML+k9ZNaegcy6NF8APhwRlyRVW6lJWkrm7fUzTbafAtwVESclU1npSfoT8APggYjYnd1WAVwM3BgRpyRZXym5LxpJGgpcBnyOfT83iIgYW/qqWldOgb46Io5vZt+fO9qn1e1J0ksRcWwz+9Zkb1RSFlrpi2b3pZH7Yl+SvhgRdyddR6HKZsoF2CzpYuDBiHgfQNIHyIw+ym2e8HFJjwEzgPXZbb2AK4AnEqsqGc9K+imZd265ffEFoMOdxdDO3BdNRMTdkj4N9CEnLyNiRmJFtaCcRuh9yNwabziNAf5PwHwy84OvJFNZMiSdS+ZUzp5kTkerB2ZFxOxECyux7AqiX2TvvlgP/I7MZyvvJVheSbkv9iXpXuDjwFJgd3ZzRMT1yVXVvLIJ9FySjiTzs3eEu4+YWQclaRXQr7MsB16Wl/5HxKbcMJf0z0nW05FkbxNogKTzkq6hoyjjvlgOdJp8KMtAz6PTfOhRAr65d6NBSRfQgZRrX/QAVkp6UtKsPV9JF9WcspxyMTMrhKTT8m2PiKdKXUshHOiApEMjYlvSdZSSpE+S+eDrmdyfXdI5EVFWZ7pIqiPzQddiSf2Ac4AXy/AD4t7AWxGxXZKAK4GBwErg3yNiV5L1Wesc6ICk1yKid9J1lIqk64GvAKuAAcANEfFIdt9zETEwyfpKSdLNwLlkTkn7PTAYWACcBTwZEd9KrrrSkrQcqIuIdyTdTubsjt+SOTOMjnoxTXuQ9MeIGCrp72SuKm/YReaPf/eESmtR2QS6pAnN7QK+HhEfLmU9SZL0AnBqRGzLns75AHBvRPxQ0vOtrTKXJtm+GAAcDPwFqIqIv0n6IJl3L9WJFlhCklZGRL/s988Cg3Ku2fh/5XQFcWdVTh+Kfhs4AjisydehlFc/AFTsmWbJrs9xOnCupO9Rfh+K7oqI3RHxDpklU/8GEBHvAu8nW1rJrZc0PPv9OjIXFe05zdc6gXK6UvQ54LcR8WzTHZKuTqCeJP1F0oCIWAqQHamfB0wH+idbWsntkHRINtBP3rNR0uGUX6BfDcyQNAXYCizNrvl9BNDcO1zrQMppyuV4YK/zz3P2faSclo6VVEVmZPqXPPuGRMR/JlBWIiQdnO8KSEk9gKMj4oUEykqUpE8Bx5EZ8NUDi/dMvVjHVjaBbmaWduU2d2xmdsAkzZX0eEe9crac5tDNzNrqCuBooEOuDe8pFzOzFnSm2/GV/ZSLpGslXZK9RV1Z6+hvJ0vJfdFI0j2S/rekE5OupVQk9ZY0U9JG4BlgsaS3stv6JFtd88o+0Mmcdz0U6JD3CCyxK4CbgI8mXUgH4L5o9BNgLtBpbpZcBPcDDwP/HBHHZu/idTSZK2dnJlpZC8pmykXSDdkrIcvqtLzWdKa3k+3NfWF7dNbb8ZXTCP2q7L8/TrSKDqCzvp1sD+6LRpIOl3SbpBclbcp+rcpu+6ek6yuxZyX9VNJgScdkvwZnb9HXYW/HV04j9PuAU4FKYG3uLjKjsnJas8N3d89yXzSS9CQwD7hnz0Vn2Zu/XAmcGRFnJ1heSXXW2/GVTaBDwy/nk8Copvsi4tXSV5SMzvp2sj24LxpJWh0Rx+/vPus4yurMjuyowyvG+e7uudwXjV6V9D/JjNDfhMyyGGRG6OtbemI5kXReRDyadB35lNUIHTJrlQBTyJy90IXGKZePJVlXKXXWt5PtwX3RSNIRwCQyfXFUdvObwCzg9oh4O6naOhJJt0TEzUnXkU85BvqLwFeBZ4Hde7ZHxKbEijIzK4JyOstlj60R8XhEvBURm/Z8JV1UR+ELaRq5LxpJKpu7WEHD2U/dst9L0lWSfizpyx35IsRyDPT5kr4j6VRJA/d8JV1UB1Kud3fPx33R6MtJF1Bis2nMx9uAkWROax0ETEuqqNaU45TL/DybIyKG59luZmWos96Or8O+dWgvEXFG0jV0ZJLOjojfJ11HKUnqDlRGxNom26sjYllCZSUie2ovEfEXSZXAvwCrI2JFspWV3HpJwyNiHo2343u1o9+Or2xG6JIuj4hfqpmbRUfE90pdU0ck6bWI6J10HaUi6XNkLix6CzgIuDIiFmf3PRcRZTMdJ+kaMme5CLidzOmKK4AhwB0RcXdy1ZWWpF7ADKCCzO34hpI5jfUI4L9HxH8kWF6zymmE/qHsv4clWkUHIGlWc7uADj0CaQf/BpwcERsk1QH3Svq3iHiI8rth9njgBOCDwKvAJ7Ij9SOA+UDZBHpErAfOyLkd3y/oBLfjK5tAj4i7sv/eknQtHcC/AJcD25psF1BX+nISVRERGwAiYpGkM4BHlbnvanm8fW20M3uz7Hckrd1z+X9EbJZUVn0hSZGxCljVUpsSl9aisgn0fMrtLXWOp4F3IuKppjskrU6gniT9XdLH98yfZ0fqp5NZJvWERCsrvfclHRQRO8mc1QFA9vS9cjsjbr6kB4FHIuK1PRuzF6INJXMl8XwyI/cOo2zm0POR9HxE1CRdhyVH0klk/ri91GT7QcDnIuJXyVRWepJ6A29ExK4m23sCn4qIuclUVnrZP2JjgcuAvsAWoBuZOfU5wNSIWJpchfmVe6D/r4i4Kek6Sq2Qt4od8e1ke3BfNHJf5Jf9494DeDcitiRdT0vK5m2UpH0+4Goa5vnapNR8SddlR2QNJHWVNFzSPWTeUpYD90Uj90UeEbEzIjZ09DCHMhqhS1oAtDonFhG/SKTAEuqsbyfbg/uikfui8yunQM/3y/pBMu9SyvaXtTO9nWxv7otG7ovOqWwCPZd/Wc0sjcoy0M3M0qhsPhQ1M0s7B7qZWUo40M3MUsKBbmaWEg50M7OU+P+lUen5NK6FcQAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"trx_sett_sum_30 PSI: 0.04025392894011593\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>trx_sett_sum_30_base</th>\n",
" <th>trx_sett_sum_30_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 14580.0]</th>\n",
" <td>0.316663</td>\n",
" <td>0.370992</td>\n",
" <td>-0.054329</td>\n",
" <td>-0.158343</td>\n",
" <td>0.008603</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(14580.0, 81058.5]</th>\n",
" <td>0.079167</td>\n",
" <td>0.104255</td>\n",
" <td>-0.025088</td>\n",
" <td>-0.275282</td>\n",
" <td>0.006906</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(81058.5, 172486.4]</th>\n",
" <td>0.079167</td>\n",
" <td>0.080814</td>\n",
" <td>-0.001647</td>\n",
" <td>-0.020593</td>\n",
" <td>0.000034</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(172486.4, 328431.8]</th>\n",
" <td>0.079161</td>\n",
" <td>0.078159</td>\n",
" <td>0.001002</td>\n",
" <td>0.012744</td>\n",
" <td>0.000013</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(328431.8, 550000.0]</th>\n",
" <td>0.079197</td>\n",
" <td>0.055209</td>\n",
" <td>0.023988</td>\n",
" <td>0.360806</td>\n",
" <td>0.008655</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(550000.0, 1000533.3]</th>\n",
" <td>0.079137</td>\n",
" <td>0.049997</td>\n",
" <td>0.029141</td>\n",
" <td>0.459226</td>\n",
" <td>0.013382</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(1000533.3, inf]</th>\n",
" <td>0.079167</td>\n",
" <td>0.072520</td>\n",
" <td>0.006647</td>\n",
" <td>0.087702</td>\n",
" <td>0.000583</td>\n",
" </tr>\n",
" <tr>\n",
" <th>NaN</th>\n",
" <td>0.208340</td>\n",
" <td>0.188053</td>\n",
" <td>0.020287</td>\n",
" <td>0.102445</td>\n",
" <td>0.002078</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" trx_sett_sum_30_base trx_sett_sum_30_prod base-prod \\\n",
"(-inf, 14580.0] 0.316663 0.370992 -0.054329 \n",
"(14580.0, 81058.5] 0.079167 0.104255 -0.025088 \n",
"(81058.5, 172486.4] 0.079167 0.080814 -0.001647 \n",
"(172486.4, 328431.8] 0.079161 0.078159 0.001002 \n",
"(328431.8, 550000.0] 0.079197 0.055209 0.023988 \n",
"(550000.0, 1000533.3] 0.079137 0.049997 0.029141 \n",
"(1000533.3, inf] 0.079167 0.072520 0.006647 \n",
"NaN 0.208340 0.188053 0.020287 \n",
"\n",
" ln(base/prod) index \n",
"(-inf, 14580.0] -0.158343 0.008603 \n",
"(14580.0, 81058.5] -0.275282 0.006906 \n",
"(81058.5, 172486.4] -0.020593 0.000034 \n",
"(172486.4, 328431.8] 0.012744 0.000013 \n",
"(328431.8, 550000.0] 0.360806 0.008655 \n",
"(550000.0, 1000533.3] 0.459226 0.013382 \n",
"(1000533.3, inf] 0.087702 0.000583 \n",
"NaN 0.102445 0.002078 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"util_non_pl PSI: 0.006531823081097055\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>util_non_pl_base</th>\n",
" <th>util_non_pl_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 0.132]</th>\n",
" <td>0.599999</td>\n",
" <td>0.615599</td>\n",
" <td>-0.015600</td>\n",
" <td>-0.025668</td>\n",
" <td>0.000400</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.132, 0.264]</th>\n",
" <td>0.100011</td>\n",
" <td>0.086880</td>\n",
" <td>0.013131</td>\n",
" <td>0.140755</td>\n",
" <td>0.001848</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.264, 0.415]</th>\n",
" <td>0.099993</td>\n",
" <td>0.087896</td>\n",
" <td>0.012097</td>\n",
" <td>0.128945</td>\n",
" <td>0.001560</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.415, 0.593]</th>\n",
" <td>0.099999</td>\n",
" <td>0.093797</td>\n",
" <td>0.006202</td>\n",
" <td>0.064024</td>\n",
" <td>0.000397</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.593, inf]</th>\n",
" <td>0.099999</td>\n",
" <td>0.115828</td>\n",
" <td>-0.015830</td>\n",
" <td>-0.146952</td>\n",
" <td>0.002326</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" util_non_pl_base util_non_pl_prod base-prod ln(base/prod) \\\n",
"(-inf, 0.132] 0.599999 0.615599 -0.015600 -0.025668 \n",
"(0.132, 0.264] 0.100011 0.086880 0.013131 0.140755 \n",
"(0.264, 0.415] 0.099993 0.087896 0.012097 0.128945 \n",
"(0.415, 0.593] 0.099999 0.093797 0.006202 0.064024 \n",
"(0.593, inf] 0.099999 0.115828 -0.015830 -0.146952 \n",
"\n",
" index \n",
"(-inf, 0.132] 0.000400 \n",
"(0.132, 0.264] 0.001848 \n",
"(0.264, 0.415] 0.001560 \n",
"(0.415, 0.593] 0.000397 \n",
"(0.593, inf] 0.002326 "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"util_pl PSI: 0.012842600447325873\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>util_pl_base</th>\n",
" <th>util_pl_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 0.213]</th>\n",
" <td>0.106040</td>\n",
" <td>0.090781</td>\n",
" <td>0.015259</td>\n",
" <td>0.155364</td>\n",
" <td>0.002371</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.213, 0.303]</th>\n",
" <td>0.105566</td>\n",
" <td>0.091502</td>\n",
" <td>0.014064</td>\n",
" <td>0.142976</td>\n",
" <td>0.002011</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.303, 0.385]</th>\n",
" <td>0.097248</td>\n",
" <td>0.087339</td>\n",
" <td>0.009909</td>\n",
" <td>0.107473</td>\n",
" <td>0.001065</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.385, 0.476]</th>\n",
" <td>0.098680</td>\n",
" <td>0.101993</td>\n",
" <td>-0.003313</td>\n",
" <td>-0.033021</td>\n",
" <td>0.000109</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.476, 0.58]</th>\n",
" <td>0.093257</td>\n",
" <td>0.093797</td>\n",
" <td>-0.000540</td>\n",
" <td>-0.005778</td>\n",
" <td>0.000003</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.58, 0.714]</th>\n",
" <td>0.109887</td>\n",
" <td>0.117140</td>\n",
" <td>-0.007253</td>\n",
" <td>-0.063914</td>\n",
" <td>0.000464</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.714, 0.862]</th>\n",
" <td>0.093107</td>\n",
" <td>0.117238</td>\n",
" <td>-0.024131</td>\n",
" <td>-0.230460</td>\n",
" <td>0.005561</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.862, 0.946]</th>\n",
" <td>0.100173</td>\n",
" <td>0.110255</td>\n",
" <td>-0.010082</td>\n",
" <td>-0.095902</td>\n",
" <td>0.000967</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.946, 0.952]</th>\n",
" <td>0.104949</td>\n",
" <td>0.099534</td>\n",
" <td>0.005415</td>\n",
" <td>0.052971</td>\n",
" <td>0.000287</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.952, inf]</th>\n",
" <td>0.091093</td>\n",
" <td>0.090420</td>\n",
" <td>0.000673</td>\n",
" <td>0.007415</td>\n",
" <td>0.000005</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" util_pl_base util_pl_prod base-prod ln(base/prod) index\n",
"(-inf, 0.213] 0.106040 0.090781 0.015259 0.155364 0.002371\n",
"(0.213, 0.303] 0.105566 0.091502 0.014064 0.142976 0.002011\n",
"(0.303, 0.385] 0.097248 0.087339 0.009909 0.107473 0.001065\n",
"(0.385, 0.476] 0.098680 0.101993 -0.003313 -0.033021 0.000109\n",
"(0.476, 0.58] 0.093257 0.093797 -0.000540 -0.005778 0.000003\n",
"(0.58, 0.714] 0.109887 0.117140 -0.007253 -0.063914 0.000464\n",
"(0.714, 0.862] 0.093107 0.117238 -0.024131 -0.230460 0.005561\n",
"(0.862, 0.946] 0.100173 0.110255 -0.010082 -0.095902 0.000967\n",
"(0.946, 0.952] 0.104949 0.099534 0.005415 0.052971 0.000287\n",
"(0.952, inf] 0.091093 0.090420 0.000673 0.007415 0.000005"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"score PSI: 0.5996344795369773\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>score_base</th>\n",
" <th>score_prod</th>\n",
" <th>base-prod</th>\n",
" <th>ln(base/prod)</th>\n",
" <th>index</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>(-inf, 0.00855]</th>\n",
" <td>0.100005</td>\n",
" <td>0.034588</td>\n",
" <td>0.065417</td>\n",
" <td>1.061714</td>\n",
" <td>0.069454</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.00855, 0.0111]</th>\n",
" <td>0.099999</td>\n",
" <td>0.044259</td>\n",
" <td>0.055739</td>\n",
" <td>0.815091</td>\n",
" <td>0.045433</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0111, 0.0145]</th>\n",
" <td>0.099999</td>\n",
" <td>0.045800</td>\n",
" <td>0.054199</td>\n",
" <td>0.780868</td>\n",
" <td>0.042322</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0145, 0.0193]</th>\n",
" <td>0.099999</td>\n",
" <td>0.049472</td>\n",
" <td>0.050527</td>\n",
" <td>0.703748</td>\n",
" <td>0.035558</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0193, 0.026]</th>\n",
" <td>0.099999</td>\n",
" <td>0.052488</td>\n",
" <td>0.047510</td>\n",
" <td>0.644567</td>\n",
" <td>0.030624</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.026, 0.0355]</th>\n",
" <td>0.099999</td>\n",
" <td>0.061373</td>\n",
" <td>0.038626</td>\n",
" <td>0.488188</td>\n",
" <td>0.018857</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0355, 0.0496]</th>\n",
" <td>0.099999</td>\n",
" <td>0.074356</td>\n",
" <td>0.025643</td>\n",
" <td>0.296297</td>\n",
" <td>0.007598</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0496, 0.0734]</th>\n",
" <td>0.099999</td>\n",
" <td>0.111042</td>\n",
" <td>-0.011043</td>\n",
" <td>-0.104749</td>\n",
" <td>0.001157</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.0734, 0.123]</th>\n",
" <td>0.099999</td>\n",
" <td>0.185562</td>\n",
" <td>-0.085563</td>\n",
" <td>-0.618229</td>\n",
" <td>0.052897</td>\n",
" </tr>\n",
" <tr>\n",
" <th>(0.123, inf]</th>\n",
" <td>0.100005</td>\n",
" <td>0.341060</td>\n",
" <td>-0.241055</td>\n",
" <td>-1.226839</td>\n",
" <td>0.295735</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" score_base score_prod base-prod ln(base/prod) index\n",
"(-inf, 0.00855] 0.100005 0.034588 0.065417 1.061714 0.069454\n",
"(0.00855, 0.0111] 0.099999 0.044259 0.055739 0.815091 0.045433\n",
"(0.0111, 0.0145] 0.099999 0.045800 0.054199 0.780868 0.042322\n",
"(0.0145, 0.0193] 0.099999 0.049472 0.050527 0.703748 0.035558\n",
"(0.0193, 0.026] 0.099999 0.052488 0.047510 0.644567 0.030624\n",
"(0.026, 0.0355] 0.099999 0.061373 0.038626 0.488188 0.018857\n",
"(0.0355, 0.0496] 0.099999 0.074356 0.025643 0.296297 0.007598\n",
"(0.0496, 0.0734] 0.099999 0.111042 -0.011043 -0.104749 0.001157\n",
"(0.0734, 0.123] 0.099999 0.185562 -0.085563 -0.618229 0.052897\n",
"(0.123, inf] 0.100005 0.341060 -0.241055 -1.226839 0.295735"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"for feat in features + ['score']:\n",
" calculate_psi(base_psi_binned, prod_psi_binned, feat)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.4"
}
},
"nbformat": 4,
"nbformat_minor": 2
}