3783 lines
497 KiB
Plaintext
3783 lines
497 KiB
Plaintext
{
|
||
"cells": [
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ee3b8d29-20bd-4c03-b366-0d16c21054c8",
|
||
"metadata": {},
|
||
"source": [
|
||
"# Clustering"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 1,
|
||
"id": "1161613a-7665-407a-9d2e-956947114767",
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"import pandas as pd\n",
|
||
"import numpy as np\n",
|
||
"import matplotlib.pyplot as plt"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 2,
|
||
"id": "ae8183d7-1ba0-443c-a5cc-14f43df82b42",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/tmp/ipykernel_9087/3355046467.py:2: DtypeWarning: Columns (1,2,3,4) have mixed types. Specify dtype option on import or set low_memory=False.\n",
|
||
" df_flows = pd.read_csv(\"flows.csv\")\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>Unnamed: 0</th>\n",
|
||
" <th>Agreement - Code</th>\n",
|
||
" <th>Company - Id</th>\n",
|
||
" <th>Company - Ultimate Parent Id</th>\n",
|
||
" <th>Registrar Account - ID</th>\n",
|
||
" <th>Registrar Account - Region</th>\n",
|
||
" <th>RegistrarAccount - Country</th>\n",
|
||
" <th>Product - Asset Type</th>\n",
|
||
" <th>Product - Strategy</th>\n",
|
||
" <th>Product - Legal Status</th>\n",
|
||
" <th>...</th>\n",
|
||
" <th>Centralisation Date</th>\n",
|
||
" <th>Quantity - Subscription</th>\n",
|
||
" <th>Quantity - Redemption</th>\n",
|
||
" <th>Quantity - NetFlows</th>\n",
|
||
" <th>Value Ccy - Subscription</th>\n",
|
||
" <th>Value Ccy - Redemption</th>\n",
|
||
" <th>Value Ccy - NetFlows</th>\n",
|
||
" <th>Value € - Subscription</th>\n",
|
||
" <th>Value € - Redemption</th>\n",
|
||
" <th>Value € - NetFlows</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>0</td>\n",
|
||
" <td>003</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>200127202</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>EQUITY</td>\n",
|
||
" <td>INVESTISSEMENT</td>\n",
|
||
" <td>SICAV</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2020-11-05</td>\n",
|
||
" <td>1636.000</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>1636.000</td>\n",
|
||
" <td>280983.00</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>280983.00</td>\n",
|
||
" <td>280983.00</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>280983.00</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>003</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>406533</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2015-03-09</td>\n",
|
||
" <td>144.690</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>144.690</td>\n",
|
||
" <td>99985.13</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>99985.13</td>\n",
|
||
" <td>99985.13</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>99985.13</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>2</td>\n",
|
||
" <td>003</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>406533</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>EQUITY</td>\n",
|
||
" <td>INVESTISSEMENT</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2016-10-26</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>-8.321</td>\n",
|
||
" <td>-8.321</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-9384.76</td>\n",
|
||
" <td>-9384.76</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-9384.76</td>\n",
|
||
" <td>-9384.76</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>3</td>\n",
|
||
" <td>003</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>406533</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>EQUITY</td>\n",
|
||
" <td>INVESTISSEMENT</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2018-10-18</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>-22.083</td>\n",
|
||
" <td>-22.083</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-25227.40</td>\n",
|
||
" <td>-25227.40</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-25227.40</td>\n",
|
||
" <td>-25227.40</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>4</td>\n",
|
||
" <td>003</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>166</td>\n",
|
||
" <td>406533</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>EQUITY</td>\n",
|
||
" <td>INVESTISSEMENT</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2019-04-08</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>-465.992</td>\n",
|
||
" <td>-465.992</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-563775.76</td>\n",
|
||
" <td>-563775.76</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-563775.76</td>\n",
|
||
" <td>-563775.76</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>...</th>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2574456</th>\n",
|
||
" <td>2574456</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2015-06-12</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>-20.000</td>\n",
|
||
" <td>-20.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-34294.40</td>\n",
|
||
" <td>-34294.40</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-34294.40</td>\n",
|
||
" <td>-34294.40</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2574457</th>\n",
|
||
" <td>2574457</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2015-09-18</td>\n",
|
||
" <td>328.726</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>328.726</td>\n",
|
||
" <td>564028.07</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>564028.07</td>\n",
|
||
" <td>564028.07</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>564028.07</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2574458</th>\n",
|
||
" <td>2574458</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2015-09-25</td>\n",
|
||
" <td>4.443</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>4.443</td>\n",
|
||
" <td>7603.66</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>7603.66</td>\n",
|
||
" <td>7603.66</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>7603.66</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2574459</th>\n",
|
||
" <td>2574459</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>FCP</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2015-11-09</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>-440.000</td>\n",
|
||
" <td>-440.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-754696.80</td>\n",
|
||
" <td>-754696.80</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>-754696.80</td>\n",
|
||
" <td>-754696.80</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2574460</th>\n",
|
||
" <td>2574460</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>LUXEMBOURG</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>SICAV</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>2016-01-11</td>\n",
|
||
" <td>3595.000</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>3595.000</td>\n",
|
||
" <td>358385.55</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>358385.55</td>\n",
|
||
" <td>358385.55</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>358385.55</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"<p>2574461 rows × 25 columns</p>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Unnamed: 0 Agreement - Code Company - Id \\\n",
|
||
"0 0 003 166 \n",
|
||
"1 1 003 166 \n",
|
||
"2 2 003 166 \n",
|
||
"3 3 003 166 \n",
|
||
"4 4 003 166 \n",
|
||
"... ... ... ... \n",
|
||
"2574456 2574456 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574457 2574457 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574458 2574458 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574459 2574459 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574460 2574460 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"\n",
|
||
" Company - Ultimate Parent Id Registrar Account - ID \\\n",
|
||
"0 166 200127202 \n",
|
||
"1 166 406533 \n",
|
||
"2 166 406533 \n",
|
||
"3 166 406533 \n",
|
||
"4 166 406533 \n",
|
||
"... ... ... \n",
|
||
"2574456 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574457 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574458 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574459 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"2574460 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"\n",
|
||
" Registrar Account - Region RegistrarAccount - Country \\\n",
|
||
"0 FRANCE FRANCE \n",
|
||
"1 FRANCE FRANCE \n",
|
||
"2 FRANCE FRANCE \n",
|
||
"3 FRANCE FRANCE \n",
|
||
"4 FRANCE FRANCE \n",
|
||
"... ... ... \n",
|
||
"2574456 LUXEMBOURG LUXEMBOURG \n",
|
||
"2574457 LUXEMBOURG LUXEMBOURG \n",
|
||
"2574458 LUXEMBOURG LUXEMBOURG \n",
|
||
"2574459 LUXEMBOURG LUXEMBOURG \n",
|
||
"2574460 LUXEMBOURG LUXEMBOURG \n",
|
||
"\n",
|
||
" Product - Asset Type Product - Strategy Product - Legal Status ... \\\n",
|
||
"0 EQUITY INVESTISSEMENT SICAV ... \n",
|
||
"1 DIVERSIFIED PATRIMOINE FCP ... \n",
|
||
"2 EQUITY INVESTISSEMENT FCP ... \n",
|
||
"3 EQUITY INVESTISSEMENT FCP ... \n",
|
||
"4 EQUITY INVESTISSEMENT FCP ... \n",
|
||
"... ... ... ... ... \n",
|
||
"2574456 FIXED INCOME SÉCURITÉ FCP ... \n",
|
||
"2574457 FIXED INCOME SÉCURITÉ FCP ... \n",
|
||
"2574458 FIXED INCOME SÉCURITÉ FCP ... \n",
|
||
"2574459 FIXED INCOME SÉCURITÉ FCP ... \n",
|
||
"2574460 FIXED INCOME SÉCURITÉ SICAV ... \n",
|
||
"\n",
|
||
" Centralisation Date Quantity - Subscription Quantity - Redemption \\\n",
|
||
"0 2020-11-05 1636.000 0.000 \n",
|
||
"1 2015-03-09 144.690 0.000 \n",
|
||
"2 2016-10-26 0.000 -8.321 \n",
|
||
"3 2018-10-18 0.000 -22.083 \n",
|
||
"4 2019-04-08 0.000 -465.992 \n",
|
||
"... ... ... ... \n",
|
||
"2574456 2015-06-12 0.000 -20.000 \n",
|
||
"2574457 2015-09-18 328.726 0.000 \n",
|
||
"2574458 2015-09-25 4.443 0.000 \n",
|
||
"2574459 2015-11-09 0.000 -440.000 \n",
|
||
"2574460 2016-01-11 3595.000 0.000 \n",
|
||
"\n",
|
||
" Quantity - NetFlows Value Ccy - Subscription Value Ccy - Redemption \\\n",
|
||
"0 1636.000 280983.00 0.00 \n",
|
||
"1 144.690 99985.13 0.00 \n",
|
||
"2 -8.321 0.00 -9384.76 \n",
|
||
"3 -22.083 0.00 -25227.40 \n",
|
||
"4 -465.992 0.00 -563775.76 \n",
|
||
"... ... ... ... \n",
|
||
"2574456 -20.000 0.00 -34294.40 \n",
|
||
"2574457 328.726 564028.07 0.00 \n",
|
||
"2574458 4.443 7603.66 0.00 \n",
|
||
"2574459 -440.000 0.00 -754696.80 \n",
|
||
"2574460 3595.000 358385.55 0.00 \n",
|
||
"\n",
|
||
" Value Ccy - NetFlows Value € - Subscription Value € - Redemption \\\n",
|
||
"0 280983.00 280983.00 0.00 \n",
|
||
"1 99985.13 99985.13 0.00 \n",
|
||
"2 -9384.76 0.00 -9384.76 \n",
|
||
"3 -25227.40 0.00 -25227.40 \n",
|
||
"4 -563775.76 0.00 -563775.76 \n",
|
||
"... ... ... ... \n",
|
||
"2574456 -34294.40 0.00 -34294.40 \n",
|
||
"2574457 564028.07 564028.07 0.00 \n",
|
||
"2574458 7603.66 7603.66 0.00 \n",
|
||
"2574459 -754696.80 0.00 -754696.80 \n",
|
||
"2574460 358385.55 358385.55 0.00 \n",
|
||
"\n",
|
||
" Value € - NetFlows \n",
|
||
"0 280983.00 \n",
|
||
"1 99985.13 \n",
|
||
"2 -9384.76 \n",
|
||
"3 -25227.40 \n",
|
||
"4 -563775.76 \n",
|
||
"... ... \n",
|
||
"2574456 -34294.40 \n",
|
||
"2574457 564028.07 \n",
|
||
"2574458 7603.66 \n",
|
||
"2574459 -754696.80 \n",
|
||
"2574460 358385.55 \n",
|
||
"\n",
|
||
"[2574461 rows x 25 columns]"
|
||
]
|
||
},
|
||
"execution_count": 2,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"import pandas as pd \n",
|
||
"df_flows = pd.read_csv(\"flows.csv\")\n",
|
||
"df_flows"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 3,
|
||
"id": "0a614aff-b7e1-4032-94d1-cb87a3cd8806",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/tmp/ipykernel_9087/223063792.py:1: DtypeWarning: Columns (2,3,4,5) have mixed types. Specify dtype option on import or set low_memory=False.\n",
|
||
" df_aum = pd.read_csv(\"AUM_repaired.csv\")\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>Unnamed: 0.1</th>\n",
|
||
" <th>Unnamed: 0</th>\n",
|
||
" <th>Agreement - Code</th>\n",
|
||
" <th>Company - Id</th>\n",
|
||
" <th>Company - Ultimate Parent Id</th>\n",
|
||
" <th>Registrar Account - ID</th>\n",
|
||
" <th>Registrar Account - Region</th>\n",
|
||
" <th>RegistrarAccount - Country</th>\n",
|
||
" <th>Product - Asset Type</th>\n",
|
||
" <th>Product - Strategy</th>\n",
|
||
" <th>...</th>\n",
|
||
" <th>Product - Is Dedie ?</th>\n",
|
||
" <th>Product - Fund</th>\n",
|
||
" <th>Product - Shareclass Type</th>\n",
|
||
" <th>Product - Shareclass Currency</th>\n",
|
||
" <th>Product - Isin</th>\n",
|
||
" <th>Centralisation Date</th>\n",
|
||
" <th>Quantity - AUM</th>\n",
|
||
" <th>Value - AUM CCY</th>\n",
|
||
" <th>Value - AUM €</th>\n",
|
||
" <th>repair_flag</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-03-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>24648.6666</td>\n",
|
||
" <td>24648.6666</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-11-30</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22413.0553</td>\n",
|
||
" <td>22413.0553</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>2</td>\n",
|
||
" <td>2</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-12-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22051.2406</td>\n",
|
||
" <td>22051.2406</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>3</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2016-03-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>21626.1173</td>\n",
|
||
" <td>21626.1173</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>4</td>\n",
|
||
" <td>4</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2016-11-30</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22489.4502</td>\n",
|
||
" <td>22489.4502</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>...</th>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5516128</th>\n",
|
||
" <td>5516128</td>\n",
|
||
" <td>4880294</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PORTFOLIO SÉCURITÉ</td>\n",
|
||
" <td>AW & AW-R</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>LU1299306321</td>\n",
|
||
" <td>2020-10-31</td>\n",
|
||
" <td>3099.000</td>\n",
|
||
" <td>318422.2500</td>\n",
|
||
" <td>318422.2500</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5516129</th>\n",
|
||
" <td>5516129</td>\n",
|
||
" <td>4880294</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PORTFOLIO SÉCURITÉ</td>\n",
|
||
" <td>AW & AW-R</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>LU1299306321</td>\n",
|
||
" <td>2020-10-31</td>\n",
|
||
" <td>3099.000</td>\n",
|
||
" <td>318422.2500</td>\n",
|
||
" <td>318422.2500</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5516130</th>\n",
|
||
" <td>5516130</td>\n",
|
||
" <td>4880295</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PORTFOLIO SÉCURITÉ</td>\n",
|
||
" <td>AW & AW-R</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>LU1299306321</td>\n",
|
||
" <td>2021-07-31</td>\n",
|
||
" <td>2835.000</td>\n",
|
||
" <td>297618.3000</td>\n",
|
||
" <td>297618.3000</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5516131</th>\n",
|
||
" <td>5516131</td>\n",
|
||
" <td>4880295</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PORTFOLIO SÉCURITÉ</td>\n",
|
||
" <td>AW & AW-R</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>LU1299306321</td>\n",
|
||
" <td>2021-07-31</td>\n",
|
||
" <td>2835.000</td>\n",
|
||
" <td>297618.3000</td>\n",
|
||
" <td>297618.3000</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5516132</th>\n",
|
||
" <td>5516132</td>\n",
|
||
" <td>4880296</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>SWITZERLAND</td>\n",
|
||
" <td>FIXED INCOME</td>\n",
|
||
" <td>SÉCURITÉ</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PORTFOLIO SÉCURITÉ</td>\n",
|
||
" <td>FW & FW-R</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>LU1792391911</td>\n",
|
||
" <td>2020-07-31</td>\n",
|
||
" <td>2916.394</td>\n",
|
||
" <td>287410.6287</td>\n",
|
||
" <td>287410.6287</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"<p>5516133 rows × 21 columns</p>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Unnamed: 0.1 Unnamed: 0 Agreement - Code Company - Id \\\n",
|
||
"0 0 0 3 166.0 \n",
|
||
"1 1 1 3 166.0 \n",
|
||
"2 2 2 3 166.0 \n",
|
||
"3 3 3 3 166.0 \n",
|
||
"4 4 4 3 166.0 \n",
|
||
"... ... ... ... ... \n",
|
||
"5516128 5516128 4880294 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516129 5516129 4880294 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516130 5516130 4880295 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516131 5516131 4880295 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516132 5516132 4880296 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"\n",
|
||
" Company - Ultimate Parent Id Registrar Account - ID \\\n",
|
||
"0 166.0 200000647 \n",
|
||
"1 166.0 200000647 \n",
|
||
"2 166.0 200000647 \n",
|
||
"3 166.0 200000647 \n",
|
||
"4 166.0 200000647 \n",
|
||
"... ... ... \n",
|
||
"5516128 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516129 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516130 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516131 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"5516132 PRIVATE CLIENT PRIVATE CLIENT \n",
|
||
"\n",
|
||
" Registrar Account - Region RegistrarAccount - Country \\\n",
|
||
"0 FRANCE FRANCE \n",
|
||
"1 FRANCE FRANCE \n",
|
||
"2 FRANCE FRANCE \n",
|
||
"3 FRANCE FRANCE \n",
|
||
"4 FRANCE FRANCE \n",
|
||
"... ... ... \n",
|
||
"5516128 SWITZERLAND SWITZERLAND \n",
|
||
"5516129 SWITZERLAND SWITZERLAND \n",
|
||
"5516130 SWITZERLAND SWITZERLAND \n",
|
||
"5516131 SWITZERLAND SWITZERLAND \n",
|
||
"5516132 SWITZERLAND SWITZERLAND \n",
|
||
"\n",
|
||
" Product - Asset Type Product - Strategy ... Product - Is Dedie ? \\\n",
|
||
"0 DIVERSIFIED PATRIMOINE ... NO \n",
|
||
"1 DIVERSIFIED PATRIMOINE ... NO \n",
|
||
"2 DIVERSIFIED PATRIMOINE ... NO \n",
|
||
"3 DIVERSIFIED PATRIMOINE ... NO \n",
|
||
"4 DIVERSIFIED PATRIMOINE ... NO \n",
|
||
"... ... ... ... ... \n",
|
||
"5516128 FIXED INCOME SÉCURITÉ ... NO \n",
|
||
"5516129 FIXED INCOME SÉCURITÉ ... NO \n",
|
||
"5516130 FIXED INCOME SÉCURITÉ ... NO \n",
|
||
"5516131 FIXED INCOME SÉCURITÉ ... NO \n",
|
||
"5516132 FIXED INCOME SÉCURITÉ ... NO \n",
|
||
"\n",
|
||
" Product - Fund Product - Shareclass Type \\\n",
|
||
"0 CARMIGNAC PATRIMOINE A \n",
|
||
"1 CARMIGNAC PATRIMOINE A \n",
|
||
"2 CARMIGNAC PATRIMOINE A \n",
|
||
"3 CARMIGNAC PATRIMOINE A \n",
|
||
"4 CARMIGNAC PATRIMOINE A \n",
|
||
"... ... ... \n",
|
||
"5516128 CARMIGNAC PORTFOLIO SÉCURITÉ AW & AW-R \n",
|
||
"5516129 CARMIGNAC PORTFOLIO SÉCURITÉ AW & AW-R \n",
|
||
"5516130 CARMIGNAC PORTFOLIO SÉCURITÉ AW & AW-R \n",
|
||
"5516131 CARMIGNAC PORTFOLIO SÉCURITÉ AW & AW-R \n",
|
||
"5516132 CARMIGNAC PORTFOLIO SÉCURITÉ FW & FW-R \n",
|
||
"\n",
|
||
" Product - Shareclass Currency Product - Isin Centralisation Date \\\n",
|
||
"0 EUR FR0010135103 2015-03-31 \n",
|
||
"1 EUR FR0010135103 2015-11-30 \n",
|
||
"2 EUR FR0010135103 2015-12-31 \n",
|
||
"3 EUR FR0010135103 2016-03-31 \n",
|
||
"4 EUR FR0010135103 2016-11-30 \n",
|
||
"... ... ... ... \n",
|
||
"5516128 EUR LU1299306321 2020-10-31 \n",
|
||
"5516129 EUR LU1299306321 2020-10-31 \n",
|
||
"5516130 EUR LU1299306321 2021-07-31 \n",
|
||
"5516131 EUR LU1299306321 2021-07-31 \n",
|
||
"5516132 EUR LU1792391911 2020-07-31 \n",
|
||
"\n",
|
||
" Quantity - AUM Value - AUM CCY Value - AUM € repair_flag \n",
|
||
"0 35.368 24648.6666 24648.6666 False \n",
|
||
"1 35.368 22413.0553 22413.0553 False \n",
|
||
"2 35.368 22051.2406 22051.2406 False \n",
|
||
"3 35.368 21626.1173 21626.1173 False \n",
|
||
"4 35.368 22489.4502 22489.4502 False \n",
|
||
"... ... ... ... ... \n",
|
||
"5516128 3099.000 318422.2500 318422.2500 False \n",
|
||
"5516129 3099.000 318422.2500 318422.2500 False \n",
|
||
"5516130 2835.000 297618.3000 297618.3000 False \n",
|
||
"5516131 2835.000 297618.3000 297618.3000 False \n",
|
||
"5516132 2916.394 287410.6287 287410.6287 False \n",
|
||
"\n",
|
||
"[5516133 rows x 21 columns]"
|
||
]
|
||
},
|
||
"execution_count": 3,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_aum = pd.read_csv(\"AUM_repaired.csv\")\n",
|
||
"df_aum"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 4,
|
||
"id": "c4fd5fc2-8977-4f22-bae3-5acb60561042",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>Registrar Account - ID</th>\n",
|
||
" <th>n_days</th>\n",
|
||
" <th>n_transactions</th>\n",
|
||
" <th>total_netflows</th>\n",
|
||
" <th>mean_flow</th>\n",
|
||
" <th>std_flow</th>\n",
|
||
" <th>total_subscription</th>\n",
|
||
" <th>total_redemption</th>\n",
|
||
" <th>churn_ratio</th>\n",
|
||
" <th>churn_flag</th>\n",
|
||
" <th>activity_score</th>\n",
|
||
" <th>flow_volatility</th>\n",
|
||
" <th>inertia_ratio</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>100000028</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>-1.092380e+02</td>\n",
|
||
" <td>-36.412667</td>\n",
|
||
" <td>49.280511</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-1.092380e+02</td>\n",
|
||
" <td>-1.092380e+08</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>1.386294</td>\n",
|
||
" <td>49.280511</td>\n",
|
||
" <td>0.998921</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>100000042</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-6.601150e+02</td>\n",
|
||
" <td>-660.115000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-6.601150e+02</td>\n",
|
||
" <td>-6.601150e+08</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>100000065</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-1.746460e+02</td>\n",
|
||
" <td>-174.646000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-1.746460e+02</td>\n",
|
||
" <td>-1.746460e+08</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>100000069</td>\n",
|
||
" <td>65</td>\n",
|
||
" <td>73</td>\n",
|
||
" <td>-7.479755e+03</td>\n",
|
||
" <td>-102.462397</td>\n",
|
||
" <td>2168.971331</td>\n",
|
||
" <td>3.332040e+04</td>\n",
|
||
" <td>-4.080016e+04</td>\n",
|
||
" <td>-1.224480e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>4.304065</td>\n",
|
||
" <td>2168.971331</td>\n",
|
||
" <td>0.976619</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>100000073</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-1.334020e+02</td>\n",
|
||
" <td>-133.402000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-1.334020e+02</td>\n",
|
||
" <td>-1.334020e+08</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>...</th>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6837</th>\n",
|
||
" <td>422905</td>\n",
|
||
" <td>208</td>\n",
|
||
" <td>212</td>\n",
|
||
" <td>-4.329218e+03</td>\n",
|
||
" <td>-20.420840</td>\n",
|
||
" <td>331.677297</td>\n",
|
||
" <td>9.699140e+03</td>\n",
|
||
" <td>-1.402836e+04</td>\n",
|
||
" <td>-1.446351e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>5.361292</td>\n",
|
||
" <td>331.677297</td>\n",
|
||
" <td>0.925180</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6838</th>\n",
|
||
" <td>422906</td>\n",
|
||
" <td>1146</td>\n",
|
||
" <td>1556</td>\n",
|
||
" <td>4.455099e+03</td>\n",
|
||
" <td>2.863174</td>\n",
|
||
" <td>201.071555</td>\n",
|
||
" <td>6.078686e+04</td>\n",
|
||
" <td>-5.633177e+04</td>\n",
|
||
" <td>-9.267095e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>7.350516</td>\n",
|
||
" <td>201.071555</td>\n",
|
||
" <td>0.587770</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6839</th>\n",
|
||
" <td>8307</td>\n",
|
||
" <td>204</td>\n",
|
||
" <td>221</td>\n",
|
||
" <td>2.168303e+04</td>\n",
|
||
" <td>98.113249</td>\n",
|
||
" <td>2217.940406</td>\n",
|
||
" <td>1.204399e+05</td>\n",
|
||
" <td>-9.875688e+04</td>\n",
|
||
" <td>-8.199681e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>5.402677</td>\n",
|
||
" <td>2217.940406</td>\n",
|
||
" <td>0.926619</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6840</th>\n",
|
||
" <td>OFF DISTRIBUTION</td>\n",
|
||
" <td>2656</td>\n",
|
||
" <td>27679</td>\n",
|
||
" <td>1.319043e+08</td>\n",
|
||
" <td>4765.499704</td>\n",
|
||
" <td>391347.475503</td>\n",
|
||
" <td>3.388942e+08</td>\n",
|
||
" <td>-2.069900e+08</td>\n",
|
||
" <td>-6.107804e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>10.228465</td>\n",
|
||
" <td>391347.475503</td>\n",
|
||
" <td>0.044604</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6841</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>2690</td>\n",
|
||
" <td>32363</td>\n",
|
||
" <td>-4.181221e+05</td>\n",
|
||
" <td>-12.919758</td>\n",
|
||
" <td>7572.830139</td>\n",
|
||
" <td>1.354277e+07</td>\n",
|
||
" <td>-1.396089e+07</td>\n",
|
||
" <td>-1.030874e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>10.384802</td>\n",
|
||
" <td>7572.830139</td>\n",
|
||
" <td>0.032374</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"<p>6842 rows × 13 columns</p>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Registrar Account - ID n_days n_transactions total_netflows \\\n",
|
||
"0 100000028 3 3 -1.092380e+02 \n",
|
||
"1 100000042 1 1 -6.601150e+02 \n",
|
||
"2 100000065 1 1 -1.746460e+02 \n",
|
||
"3 100000069 65 73 -7.479755e+03 \n",
|
||
"4 100000073 1 1 -1.334020e+02 \n",
|
||
"... ... ... ... ... \n",
|
||
"6837 422905 208 212 -4.329218e+03 \n",
|
||
"6838 422906 1146 1556 4.455099e+03 \n",
|
||
"6839 8307 204 221 2.168303e+04 \n",
|
||
"6840 OFF DISTRIBUTION 2656 27679 1.319043e+08 \n",
|
||
"6841 PRIVATE CLIENT 2690 32363 -4.181221e+05 \n",
|
||
"\n",
|
||
" mean_flow std_flow total_subscription total_redemption \\\n",
|
||
"0 -36.412667 49.280511 0.000000e+00 -1.092380e+02 \n",
|
||
"1 -660.115000 NaN 0.000000e+00 -6.601150e+02 \n",
|
||
"2 -174.646000 NaN 0.000000e+00 -1.746460e+02 \n",
|
||
"3 -102.462397 2168.971331 3.332040e+04 -4.080016e+04 \n",
|
||
"4 -133.402000 NaN 0.000000e+00 -1.334020e+02 \n",
|
||
"... ... ... ... ... \n",
|
||
"6837 -20.420840 331.677297 9.699140e+03 -1.402836e+04 \n",
|
||
"6838 2.863174 201.071555 6.078686e+04 -5.633177e+04 \n",
|
||
"6839 98.113249 2217.940406 1.204399e+05 -9.875688e+04 \n",
|
||
"6840 4765.499704 391347.475503 3.388942e+08 -2.069900e+08 \n",
|
||
"6841 -12.919758 7572.830139 1.354277e+07 -1.396089e+07 \n",
|
||
"\n",
|
||
" churn_ratio churn_flag activity_score flow_volatility inertia_ratio \n",
|
||
"0 -1.092380e+08 0 1.386294 49.280511 0.998921 \n",
|
||
"1 -6.601150e+08 0 0.693147 0.000000 0.999640 \n",
|
||
"2 -1.746460e+08 0 0.693147 0.000000 0.999640 \n",
|
||
"3 -1.224480e+00 0 4.304065 2168.971331 0.976619 \n",
|
||
"4 -1.334020e+08 0 0.693147 0.000000 0.999640 \n",
|
||
"... ... ... ... ... ... \n",
|
||
"6837 -1.446351e+00 0 5.361292 331.677297 0.925180 \n",
|
||
"6838 -9.267095e-01 0 7.350516 201.071555 0.587770 \n",
|
||
"6839 -8.199681e-01 0 5.402677 2217.940406 0.926619 \n",
|
||
"6840 -6.107804e-01 0 10.228465 391347.475503 0.044604 \n",
|
||
"6841 -1.030874e+00 0 10.384802 7572.830139 0.032374 \n",
|
||
"\n",
|
||
"[6842 rows x 13 columns]"
|
||
]
|
||
},
|
||
"execution_count": 4,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_client = pd.read_csv(\"client_behavior_clean.csv\")\n",
|
||
"df_client"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 6,
|
||
"id": "e2233e16-958e-4973-aed4-6c9f2cbe8397",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Flows - AUM: (2574461, 25) (5516133, 21)\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"print (\"Flows - AUM:\", df_flows.shape , df_aum.shape)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 5,
|
||
"id": "9bca9076-a961-4004-9b5c-7b55191e2c6e",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Flows - AUM - Client shape: (2574461, 25) (5516133, 21) (6842, 13)\n",
|
||
"Flows Columns: Index(['Unnamed: 0', 'Agreement - Code', 'Company - Id',\n",
|
||
" 'Company - Ultimate Parent Id', 'Registrar Account - ID',\n",
|
||
" 'Registrar Account - Region', 'RegistrarAccount - Country',\n",
|
||
" 'Product - Asset Type', 'Product - Strategy', 'Product - Legal Status',\n",
|
||
" 'Product - Is Dedie ?', 'Product - Fund', 'Product - Shareclass Type',\n",
|
||
" 'Product - Shareclass Currency', 'Product - Isin',\n",
|
||
" 'Centralisation Date', 'Quantity - Subscription',\n",
|
||
" 'Quantity - Redemption', 'Quantity - NetFlows',\n",
|
||
" 'Value Ccy - Subscription', 'Value Ccy - Redemption',\n",
|
||
" 'Value Ccy - NetFlows', 'Value € - Subscription',\n",
|
||
" 'Value € - Redemption', 'Value € - NetFlows'],\n",
|
||
" dtype='object')\n",
|
||
"AUM Columns: Index(['Unnamed: 0.1', 'Unnamed: 0', 'Agreement - Code', 'Company - Id',\n",
|
||
" 'Company - Ultimate Parent Id', 'Registrar Account - ID',\n",
|
||
" 'Registrar Account - Region', 'RegistrarAccount - Country',\n",
|
||
" 'Product - Asset Type', 'Product - Strategy', 'Product - Legal Status',\n",
|
||
" 'Product - Is Dedie ?', 'Product - Fund', 'Product - Shareclass Type',\n",
|
||
" 'Product - Shareclass Currency', 'Product - Isin',\n",
|
||
" 'Centralisation Date', 'Quantity - AUM', 'Value - AUM CCY',\n",
|
||
" 'Value - AUM €', 'repair_flag'],\n",
|
||
" dtype='object')\n",
|
||
"Client Columns: Index(['Registrar Account - ID', 'n_days', 'n_transactions', 'total_netflows',\n",
|
||
" 'mean_flow', 'std_flow', 'total_subscription', 'total_redemption',\n",
|
||
" 'churn_ratio', 'churn_flag', 'activity_score', 'flow_volatility',\n",
|
||
" 'inertia_ratio'],\n",
|
||
" dtype='object')\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"print (\"Flows - AUM - Client shape:\", df_flows.shape , df_aum.shape, df_client.shape)\n",
|
||
"print (\"Flows Columns:\" , df_flows.columns )\n",
|
||
"print (\"AUM Columns:\" , df_aum.columns )\n",
|
||
"print (\"Client Columns:\" , df_client.columns )"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 6,
|
||
"id": "c2a1a52f-b0b9-4ecd-bf9f-2f9759e0cd24",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>Unnamed: 0.1</th>\n",
|
||
" <th>Unnamed: 0</th>\n",
|
||
" <th>Agreement - Code</th>\n",
|
||
" <th>Company - Id</th>\n",
|
||
" <th>Company - Ultimate Parent Id</th>\n",
|
||
" <th>Registrar Account - ID</th>\n",
|
||
" <th>Registrar Account - Region</th>\n",
|
||
" <th>RegistrarAccount - Country</th>\n",
|
||
" <th>Product - Asset Type</th>\n",
|
||
" <th>Product - Strategy</th>\n",
|
||
" <th>...</th>\n",
|
||
" <th>Product - Is Dedie ?</th>\n",
|
||
" <th>Product - Fund</th>\n",
|
||
" <th>Product - Shareclass Type</th>\n",
|
||
" <th>Product - Shareclass Currency</th>\n",
|
||
" <th>Product - Isin</th>\n",
|
||
" <th>Centralisation Date</th>\n",
|
||
" <th>Quantity - AUM</th>\n",
|
||
" <th>Value - AUM CCY</th>\n",
|
||
" <th>Value - AUM €</th>\n",
|
||
" <th>repair_flag</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-03-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>24648.6666</td>\n",
|
||
" <td>24648.6666</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-11-30</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22413.0553</td>\n",
|
||
" <td>22413.0553</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>2</td>\n",
|
||
" <td>2</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2015-12-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22051.2406</td>\n",
|
||
" <td>22051.2406</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>3</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2016-03-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>21626.1173</td>\n",
|
||
" <td>21626.1173</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>4</td>\n",
|
||
" <td>4</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2016-11-30</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22489.4502</td>\n",
|
||
" <td>22489.4502</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5</th>\n",
|
||
" <td>5</td>\n",
|
||
" <td>5</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2017-06-30</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>23225.4582</td>\n",
|
||
" <td>23225.4582</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6</th>\n",
|
||
" <td>6</td>\n",
|
||
" <td>6</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2017-08-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22964.0887</td>\n",
|
||
" <td>22964.0887</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>7</th>\n",
|
||
" <td>7</td>\n",
|
||
" <td>7</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2018-03-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22692.4625</td>\n",
|
||
" <td>22692.4625</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>8</th>\n",
|
||
" <td>8</td>\n",
|
||
" <td>8</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2018-05-31</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>22749.7586</td>\n",
|
||
" <td>22749.7586</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>9</th>\n",
|
||
" <td>9</td>\n",
|
||
" <td>9</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>166.0</td>\n",
|
||
" <td>200000647</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>FRANCE</td>\n",
|
||
" <td>DIVERSIFIED</td>\n",
|
||
" <td>PATRIMOINE</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>NO</td>\n",
|
||
" <td>CARMIGNAC PATRIMOINE</td>\n",
|
||
" <td>A</td>\n",
|
||
" <td>EUR</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>2019-02-28</td>\n",
|
||
" <td>35.368</td>\n",
|
||
" <td>20875.9620</td>\n",
|
||
" <td>20875.9620</td>\n",
|
||
" <td>False</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"<p>10 rows × 21 columns</p>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Unnamed: 0.1 Unnamed: 0 Agreement - Code Company - Id \\\n",
|
||
"0 0 0 3 166.0 \n",
|
||
"1 1 1 3 166.0 \n",
|
||
"2 2 2 3 166.0 \n",
|
||
"3 3 3 3 166.0 \n",
|
||
"4 4 4 3 166.0 \n",
|
||
"5 5 5 3 166.0 \n",
|
||
"6 6 6 3 166.0 \n",
|
||
"7 7 7 3 166.0 \n",
|
||
"8 8 8 3 166.0 \n",
|
||
"9 9 9 3 166.0 \n",
|
||
"\n",
|
||
" Company - Ultimate Parent Id Registrar Account - ID \\\n",
|
||
"0 166.0 200000647 \n",
|
||
"1 166.0 200000647 \n",
|
||
"2 166.0 200000647 \n",
|
||
"3 166.0 200000647 \n",
|
||
"4 166.0 200000647 \n",
|
||
"5 166.0 200000647 \n",
|
||
"6 166.0 200000647 \n",
|
||
"7 166.0 200000647 \n",
|
||
"8 166.0 200000647 \n",
|
||
"9 166.0 200000647 \n",
|
||
"\n",
|
||
" Registrar Account - Region RegistrarAccount - Country Product - Asset Type \\\n",
|
||
"0 FRANCE FRANCE DIVERSIFIED \n",
|
||
"1 FRANCE FRANCE DIVERSIFIED \n",
|
||
"2 FRANCE FRANCE DIVERSIFIED \n",
|
||
"3 FRANCE FRANCE DIVERSIFIED \n",
|
||
"4 FRANCE FRANCE DIVERSIFIED \n",
|
||
"5 FRANCE FRANCE DIVERSIFIED \n",
|
||
"6 FRANCE FRANCE DIVERSIFIED \n",
|
||
"7 FRANCE FRANCE DIVERSIFIED \n",
|
||
"8 FRANCE FRANCE DIVERSIFIED \n",
|
||
"9 FRANCE FRANCE DIVERSIFIED \n",
|
||
"\n",
|
||
" Product - Strategy ... Product - Is Dedie ? Product - Fund \\\n",
|
||
"0 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"1 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"2 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"3 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"4 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"5 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"6 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"7 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"8 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"9 PATRIMOINE ... NO CARMIGNAC PATRIMOINE \n",
|
||
"\n",
|
||
" Product - Shareclass Type Product - Shareclass Currency Product - Isin \\\n",
|
||
"0 A EUR FR0010135103 \n",
|
||
"1 A EUR FR0010135103 \n",
|
||
"2 A EUR FR0010135103 \n",
|
||
"3 A EUR FR0010135103 \n",
|
||
"4 A EUR FR0010135103 \n",
|
||
"5 A EUR FR0010135103 \n",
|
||
"6 A EUR FR0010135103 \n",
|
||
"7 A EUR FR0010135103 \n",
|
||
"8 A EUR FR0010135103 \n",
|
||
"9 A EUR FR0010135103 \n",
|
||
"\n",
|
||
" Centralisation Date Quantity - AUM Value - AUM CCY Value - AUM € \\\n",
|
||
"0 2015-03-31 35.368 24648.6666 24648.6666 \n",
|
||
"1 2015-11-30 35.368 22413.0553 22413.0553 \n",
|
||
"2 2015-12-31 35.368 22051.2406 22051.2406 \n",
|
||
"3 2016-03-31 35.368 21626.1173 21626.1173 \n",
|
||
"4 2016-11-30 35.368 22489.4502 22489.4502 \n",
|
||
"5 2017-06-30 35.368 23225.4582 23225.4582 \n",
|
||
"6 2017-08-31 35.368 22964.0887 22964.0887 \n",
|
||
"7 2018-03-31 35.368 22692.4625 22692.4625 \n",
|
||
"8 2018-05-31 35.368 22749.7586 22749.7586 \n",
|
||
"9 2019-02-28 35.368 20875.9620 20875.9620 \n",
|
||
"\n",
|
||
" repair_flag \n",
|
||
"0 False \n",
|
||
"1 False \n",
|
||
"2 False \n",
|
||
"3 False \n",
|
||
"4 False \n",
|
||
"5 False \n",
|
||
"6 False \n",
|
||
"7 False \n",
|
||
"8 False \n",
|
||
"9 False \n",
|
||
"\n",
|
||
"[10 rows x 21 columns]"
|
||
]
|
||
},
|
||
"execution_count": 6,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_aum.head(10)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 7,
|
||
"id": "fc435518-e8a1-4ad5-bb72-e482e3cc6d75",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>Registrar Account - ID</th>\n",
|
||
" <th>n_days</th>\n",
|
||
" <th>n_transactions</th>\n",
|
||
" <th>total_netflows</th>\n",
|
||
" <th>mean_flow</th>\n",
|
||
" <th>std_flow</th>\n",
|
||
" <th>total_subscription</th>\n",
|
||
" <th>total_redemption</th>\n",
|
||
" <th>churn_ratio</th>\n",
|
||
" <th>churn_flag</th>\n",
|
||
" <th>activity_score</th>\n",
|
||
" <th>flow_volatility</th>\n",
|
||
" <th>inertia_ratio</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>6832</th>\n",
|
||
" <td>422876</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-1.229300e+01</td>\n",
|
||
" <td>-12.293000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-1.229300e+01</td>\n",
|
||
" <td>-1.229300e+07</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6833</th>\n",
|
||
" <td>422877</td>\n",
|
||
" <td>9</td>\n",
|
||
" <td>10</td>\n",
|
||
" <td>-5.860000e+02</td>\n",
|
||
" <td>-58.600000</td>\n",
|
||
" <td>34788.810510</td>\n",
|
||
" <td>1.096470e+05</td>\n",
|
||
" <td>-1.102330e+05</td>\n",
|
||
" <td>-1.005344e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>2.397895</td>\n",
|
||
" <td>34788.810510</td>\n",
|
||
" <td>0.996763</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6834</th>\n",
|
||
" <td>422884</td>\n",
|
||
" <td>8</td>\n",
|
||
" <td>8</td>\n",
|
||
" <td>1.807000e+04</td>\n",
|
||
" <td>2258.750000</td>\n",
|
||
" <td>7296.392333</td>\n",
|
||
" <td>2.044500e+04</td>\n",
|
||
" <td>-2.375000e+03</td>\n",
|
||
" <td>-1.161653e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>2.197225</td>\n",
|
||
" <td>7296.392333</td>\n",
|
||
" <td>0.997122</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6835</th>\n",
|
||
" <td>422891</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-5.358000e+01</td>\n",
|
||
" <td>-53.580000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-5.358000e+01</td>\n",
|
||
" <td>-5.358000e+07</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6836</th>\n",
|
||
" <td>422895</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>-1.811520e+02</td>\n",
|
||
" <td>-181.152000</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>-1.811520e+02</td>\n",
|
||
" <td>-1.811520e+08</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.693147</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.999640</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6837</th>\n",
|
||
" <td>422905</td>\n",
|
||
" <td>208</td>\n",
|
||
" <td>212</td>\n",
|
||
" <td>-4.329218e+03</td>\n",
|
||
" <td>-20.420840</td>\n",
|
||
" <td>331.677297</td>\n",
|
||
" <td>9.699140e+03</td>\n",
|
||
" <td>-1.402836e+04</td>\n",
|
||
" <td>-1.446351e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>5.361292</td>\n",
|
||
" <td>331.677297</td>\n",
|
||
" <td>0.925180</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6838</th>\n",
|
||
" <td>422906</td>\n",
|
||
" <td>1146</td>\n",
|
||
" <td>1556</td>\n",
|
||
" <td>4.455099e+03</td>\n",
|
||
" <td>2.863174</td>\n",
|
||
" <td>201.071555</td>\n",
|
||
" <td>6.078686e+04</td>\n",
|
||
" <td>-5.633177e+04</td>\n",
|
||
" <td>-9.267095e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>7.350516</td>\n",
|
||
" <td>201.071555</td>\n",
|
||
" <td>0.587770</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6839</th>\n",
|
||
" <td>8307</td>\n",
|
||
" <td>204</td>\n",
|
||
" <td>221</td>\n",
|
||
" <td>2.168303e+04</td>\n",
|
||
" <td>98.113249</td>\n",
|
||
" <td>2217.940406</td>\n",
|
||
" <td>1.204399e+05</td>\n",
|
||
" <td>-9.875688e+04</td>\n",
|
||
" <td>-8.199681e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>5.402677</td>\n",
|
||
" <td>2217.940406</td>\n",
|
||
" <td>0.926619</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6840</th>\n",
|
||
" <td>OFF DISTRIBUTION</td>\n",
|
||
" <td>2656</td>\n",
|
||
" <td>27679</td>\n",
|
||
" <td>1.319043e+08</td>\n",
|
||
" <td>4765.499704</td>\n",
|
||
" <td>391347.475503</td>\n",
|
||
" <td>3.388942e+08</td>\n",
|
||
" <td>-2.069900e+08</td>\n",
|
||
" <td>-6.107804e-01</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>10.228465</td>\n",
|
||
" <td>391347.475503</td>\n",
|
||
" <td>0.044604</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6841</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>2690</td>\n",
|
||
" <td>32363</td>\n",
|
||
" <td>-4.181221e+05</td>\n",
|
||
" <td>-12.919758</td>\n",
|
||
" <td>7572.830139</td>\n",
|
||
" <td>1.354277e+07</td>\n",
|
||
" <td>-1.396089e+07</td>\n",
|
||
" <td>-1.030874e+00</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>10.384802</td>\n",
|
||
" <td>7572.830139</td>\n",
|
||
" <td>0.032374</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Registrar Account - ID n_days n_transactions total_netflows \\\n",
|
||
"6832 422876 1 1 -1.229300e+01 \n",
|
||
"6833 422877 9 10 -5.860000e+02 \n",
|
||
"6834 422884 8 8 1.807000e+04 \n",
|
||
"6835 422891 1 1 -5.358000e+01 \n",
|
||
"6836 422895 1 1 -1.811520e+02 \n",
|
||
"6837 422905 208 212 -4.329218e+03 \n",
|
||
"6838 422906 1146 1556 4.455099e+03 \n",
|
||
"6839 8307 204 221 2.168303e+04 \n",
|
||
"6840 OFF DISTRIBUTION 2656 27679 1.319043e+08 \n",
|
||
"6841 PRIVATE CLIENT 2690 32363 -4.181221e+05 \n",
|
||
"\n",
|
||
" mean_flow std_flow total_subscription total_redemption \\\n",
|
||
"6832 -12.293000 NaN 0.000000e+00 -1.229300e+01 \n",
|
||
"6833 -58.600000 34788.810510 1.096470e+05 -1.102330e+05 \n",
|
||
"6834 2258.750000 7296.392333 2.044500e+04 -2.375000e+03 \n",
|
||
"6835 -53.580000 NaN 0.000000e+00 -5.358000e+01 \n",
|
||
"6836 -181.152000 NaN 0.000000e+00 -1.811520e+02 \n",
|
||
"6837 -20.420840 331.677297 9.699140e+03 -1.402836e+04 \n",
|
||
"6838 2.863174 201.071555 6.078686e+04 -5.633177e+04 \n",
|
||
"6839 98.113249 2217.940406 1.204399e+05 -9.875688e+04 \n",
|
||
"6840 4765.499704 391347.475503 3.388942e+08 -2.069900e+08 \n",
|
||
"6841 -12.919758 7572.830139 1.354277e+07 -1.396089e+07 \n",
|
||
"\n",
|
||
" churn_ratio churn_flag activity_score flow_volatility inertia_ratio \n",
|
||
"6832 -1.229300e+07 0 0.693147 0.000000 0.999640 \n",
|
||
"6833 -1.005344e+00 0 2.397895 34788.810510 0.996763 \n",
|
||
"6834 -1.161653e-01 0 2.197225 7296.392333 0.997122 \n",
|
||
"6835 -5.358000e+07 0 0.693147 0.000000 0.999640 \n",
|
||
"6836 -1.811520e+08 0 0.693147 0.000000 0.999640 \n",
|
||
"6837 -1.446351e+00 0 5.361292 331.677297 0.925180 \n",
|
||
"6838 -9.267095e-01 0 7.350516 201.071555 0.587770 \n",
|
||
"6839 -8.199681e-01 0 5.402677 2217.940406 0.926619 \n",
|
||
"6840 -6.107804e-01 0 10.228465 391347.475503 0.044604 \n",
|
||
"6841 -1.030874e+00 0 10.384802 7572.830139 0.032374 "
|
||
]
|
||
},
|
||
"execution_count": 7,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_client.tail(10)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 10,
|
||
"id": "c61af497-1950-4a94-9b6d-51548466e133",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"491"
|
||
]
|
||
},
|
||
"execution_count": 10,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_aum['Product - Isin'].nunique()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 12,
|
||
"id": "7f0dd0a7-5800-46b1-a10c-682d85377385",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"Registrar Account - ID 15491\n",
|
||
"Product - Isin 491\n",
|
||
"dtype: int64"
|
||
]
|
||
},
|
||
"execution_count": 12,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_aum[['Registrar Account - ID','Product - Isin']].nunique()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 13,
|
||
"id": "446e199c-b6a0-47e3-ac3e-50791549353e",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>Registrar Account - ID</th>\n",
|
||
" <th>Product - Isin</th>\n",
|
||
" <th>n_snapshots</th>\n",
|
||
" <th>first_date</th>\n",
|
||
" <th>last_date</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>FR0010135103</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>2021-08-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>FR0010147603</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>2021-08-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>FR0010148981</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>2021-08-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>FR0010148999</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>2021-08-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>FR0010149096</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>2021-08-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>...</th>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" <td>...</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>79654</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LU2715954330</td>\n",
|
||
" <td>3</td>\n",
|
||
" <td>2025-08-31</td>\n",
|
||
" <td>2025-10-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>79655</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LU2715954504</td>\n",
|
||
" <td>4</td>\n",
|
||
" <td>2025-07-31</td>\n",
|
||
" <td>2025-10-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>79656</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LU2799473124</td>\n",
|
||
" <td>13</td>\n",
|
||
" <td>2024-10-31</td>\n",
|
||
" <td>2025-10-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>79657</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LU2809794220</td>\n",
|
||
" <td>12</td>\n",
|
||
" <td>2024-11-30</td>\n",
|
||
" <td>2025-10-31</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>79658</th>\n",
|
||
" <td>PRIVATE CLIENT</td>\n",
|
||
" <td>LU2809794576</td>\n",
|
||
" <td>2</td>\n",
|
||
" <td>2025-09-30</td>\n",
|
||
" <td>2025-10-31</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"<p>79659 rows × 5 columns</p>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Registrar Account - ID Product - Isin n_snapshots first_date \\\n",
|
||
"0 7905 FR0010135103 80 2015-01-31 \n",
|
||
"1 7905 FR0010147603 80 2015-01-31 \n",
|
||
"2 7905 FR0010148981 80 2015-01-31 \n",
|
||
"3 7905 FR0010148999 80 2015-01-31 \n",
|
||
"4 7905 FR0010149096 80 2015-01-31 \n",
|
||
"... ... ... ... ... \n",
|
||
"79654 PRIVATE CLIENT LU2715954330 3 2025-08-31 \n",
|
||
"79655 PRIVATE CLIENT LU2715954504 4 2025-07-31 \n",
|
||
"79656 PRIVATE CLIENT LU2799473124 13 2024-10-31 \n",
|
||
"79657 PRIVATE CLIENT LU2809794220 12 2024-11-30 \n",
|
||
"79658 PRIVATE CLIENT LU2809794576 2 2025-09-30 \n",
|
||
"\n",
|
||
" last_date \n",
|
||
"0 2021-08-31 \n",
|
||
"1 2021-08-31 \n",
|
||
"2 2021-08-31 \n",
|
||
"3 2021-08-31 \n",
|
||
"4 2021-08-31 \n",
|
||
"... ... \n",
|
||
"79654 2025-10-31 \n",
|
||
"79655 2025-10-31 \n",
|
||
"79656 2025-10-31 \n",
|
||
"79657 2025-10-31 \n",
|
||
"79658 2025-10-31 \n",
|
||
"\n",
|
||
"[79659 rows x 5 columns]"
|
||
]
|
||
},
|
||
"execution_count": 13,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df= df_aum.copy()\n",
|
||
"snap_count = (\n",
|
||
" df.groupby([\"Registrar Account - ID\", \"Product - Isin\"])\n",
|
||
" .agg(\n",
|
||
" n_snapshots=(\"Centralisation Date\", \"nunique\"),\n",
|
||
" first_date=(\"Centralisation Date\", \"min\"),\n",
|
||
" last_date=(\"Centralisation Date\", \"max\")\n",
|
||
" )\n",
|
||
" .reset_index()\n",
|
||
")\n",
|
||
"\n",
|
||
"snap_count\n",
|
||
"#pour aum y a a peut pres 12 valeurs par an (certain en ont moins), mais bcp plus de lignes dans df aum que df flows rev c est a cause de quoi"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "514db9b1-5373-4412-8d6b-25abcbc0d741",
|
||
"metadata": {},
|
||
"source": [
|
||
"Pour chaque couple (Account, ISIN) :\n",
|
||
"n_snapshots : combien de fois l’AUM est observé\n",
|
||
"first_date, last_date : période couverte\n",
|
||
"\n",
|
||
"Avant clustering, je veux savoir si :\n",
|
||
"- certains portefeuilles sont très fragmentés,\n",
|
||
"- certains comptes n’ont que 1 ou 2 observations (peu fiables),\n",
|
||
"- la couverture temporelle est stable"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "000095ec-b983-4169-be11-28bbb68fc295",
|
||
"metadata": {},
|
||
"source": [
|
||
"## debut etude clustering"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 14,
|
||
"id": "e77598d7-ee1e-4932-91e5-71a190d04e7a",
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"# Parse dates\n",
|
||
"df_flows[\"Centralisation Date\"] = pd.to_datetime(df_flows[\"Centralisation Date\"], errors=\"coerce\")\n",
|
||
"df_aum[\"Centralisation Date\"] = pd.to_datetime(df_aum[\"Centralisation Date\"], errors=\"coerce\")\n",
|
||
"\n",
|
||
"ID_COL = \"Registrar Account - ID\"\n",
|
||
"FLOW_COL = \"Quantity - NetFlows\"\n",
|
||
"AUM_COL = \"Quantity - AUM\"\n",
|
||
"\n",
|
||
"# Month key\n",
|
||
"df_flows[\"month\"] = df_flows[\"Centralisation Date\"].dt.to_period(\"M\").dt.to_timestamp(\"M\")\n",
|
||
"df_aum[\"month\"] = df_aum[\"Centralisation Date\"].dt.to_period(\"M\").dt.to_timestamp(\"M\")\n",
|
||
"# Flows sont journaliers, AUM est mensuel → il faut une granularité commune."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 12,
|
||
"id": "7d933eba-1a21-486f-9518-34b43a6fee51",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"(968429, 6)\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>Registrar Account - ID</th>\n",
|
||
" <th>month</th>\n",
|
||
" <th>aum_qty</th>\n",
|
||
" <th>net_flow_qty</th>\n",
|
||
" <th>gross_flow_qty</th>\n",
|
||
" <th>n_tx</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>2015-01-31</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>2015-02-28</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>2015-03-31</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>2015-04-30</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>2015-05-31</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Registrar Account - ID month aum_qty net_flow_qty gross_flow_qty \\\n",
|
||
"0 7905 2015-01-31 0.0 0.0 0.0 \n",
|
||
"1 7905 2015-02-28 0.0 0.0 0.0 \n",
|
||
"2 7905 2015-03-31 0.0 0.0 0.0 \n",
|
||
"3 7905 2015-04-30 0.0 0.0 0.0 \n",
|
||
"4 7905 2015-05-31 0.0 0.0 0.0 \n",
|
||
"\n",
|
||
" n_tx \n",
|
||
"0 0 \n",
|
||
"1 0 \n",
|
||
"2 0 \n",
|
||
"3 0 \n",
|
||
"4 0 "
|
||
]
|
||
},
|
||
"execution_count": 12,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"# 1) Monthly aggregation for FLOWS : je fais mon etude mensuel parce que aum valeur mensuel \n",
|
||
"\n",
|
||
"ID_COL = \"Registrar Account - ID\"\n",
|
||
"FLOW_COL = \"Quantity - NetFlows\"\n",
|
||
"AUM_COL = \"Quantity - AUM\"\n",
|
||
"\n",
|
||
"df_flows_m = (\n",
|
||
" df_flows\n",
|
||
" .dropna(subset=[ID_COL, \"month\", FLOW_COL])\n",
|
||
" .assign(gross_flow_qty=lambda x: x[FLOW_COL].abs()) # absolute quantity moved\n",
|
||
" .groupby([ID_COL, \"month\"], as_index=False)\n",
|
||
" .agg(\n",
|
||
" net_flow_qty=(FLOW_COL, \"sum\"), # net quantity change over the month\n",
|
||
" gross_flow_qty=(\"gross_flow_qty\", \"sum\"), # total traded quantity (activity intensity)\n",
|
||
" n_tx=(FLOW_COL, \"size\"), # number of transactions\n",
|
||
" )\n",
|
||
")\n",
|
||
"\n",
|
||
"# 2) Monthly aggregation for AUM (client-month holdings) ---\n",
|
||
"df_aum_m = (\n",
|
||
" df_aum\n",
|
||
" .dropna(subset=[ID_COL, \"month\", AUM_COL])\n",
|
||
" .groupby([ID_COL, \"month\"], as_index=False)\n",
|
||
" .agg(aum_qty=(AUM_COL, \"sum\")) # total held quantity across ISINs\n",
|
||
")\n",
|
||
"\n",
|
||
"df_month = df_aum_m.merge(df_flows_m, on=[ID_COL, \"month\"], how=\"left\")\n",
|
||
"\n",
|
||
"# 4) Months without transactions => flows are 0 ---\n",
|
||
"df_month[\"net_flow_qty\"] = df_month[\"net_flow_qty\"].fillna(0.0)\n",
|
||
"df_month[\"gross_flow_qty\"] = df_month[\"gross_flow_qty\"].fillna(0.0)\n",
|
||
"df_month[\"n_tx\"] = df_month[\"n_tx\"].fillna(0).astype(int)\n",
|
||
"\n",
|
||
"print(df_month.shape)\n",
|
||
"df_month.head()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 13,
|
||
"id": "887c9159-a86f-487e-8dfe-617132a73fbe",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>Registrar Account - ID</th>\n",
|
||
" <th>n_months</th>\n",
|
||
" <th>n_active_months</th>\n",
|
||
" <th>flow_freq</th>\n",
|
||
" <th>aum_qty_mean</th>\n",
|
||
" <th>aum_qty_median</th>\n",
|
||
" <th>net_flow_qty_sum</th>\n",
|
||
" <th>gross_flow_qty_sum</th>\n",
|
||
" <th>gross_flow_qty_mean</th>\n",
|
||
" <th>net_flow_qty_vol</th>\n",
|
||
" <th>rel_intensity</th>\n",
|
||
" <th>netflow_to_aum</th>\n",
|
||
" <th>n_tx_total</th>\n",
|
||
" <th>log_aum_qty_mean</th>\n",
|
||
" <th>log_gross_flow_qty_mean</th>\n",
|
||
" <th>gross_flow_to_aum</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0</th>\n",
|
||
" <td>7905</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1</th>\n",
|
||
" <td>7912</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2</th>\n",
|
||
" <td>7962</td>\n",
|
||
" <td>80</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3</th>\n",
|
||
" <td>8307</td>\n",
|
||
" <td>130</td>\n",
|
||
" <td>77</td>\n",
|
||
" <td>0.592308</td>\n",
|
||
" <td>21568.649523</td>\n",
|
||
" <td>20698.97</td>\n",
|
||
" <td>27252.124</td>\n",
|
||
" <td>177077.31</td>\n",
|
||
" <td>1362.133154</td>\n",
|
||
" <td>3508.455222</td>\n",
|
||
" <td>0.056109</td>\n",
|
||
" <td>0.011473</td>\n",
|
||
" <td>161</td>\n",
|
||
" <td>9.979042</td>\n",
|
||
" <td>7.217541</td>\n",
|
||
" <td>8.20994</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4</th>\n",
|
||
" <td>8354</td>\n",
|
||
" <td>64</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000</td>\n",
|
||
" <td>0.00</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.00000</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" Registrar Account - ID n_months n_active_months flow_freq aum_qty_mean \\\n",
|
||
"0 7905 80 0 0.000000 0.000000 \n",
|
||
"1 7912 80 0 0.000000 0.000000 \n",
|
||
"2 7962 80 0 0.000000 0.000000 \n",
|
||
"3 8307 130 77 0.592308 21568.649523 \n",
|
||
"4 8354 64 0 0.000000 0.000000 \n",
|
||
"\n",
|
||
" aum_qty_median net_flow_qty_sum gross_flow_qty_sum gross_flow_qty_mean \\\n",
|
||
"0 0.00 0.000 0.00 0.000000 \n",
|
||
"1 0.00 0.000 0.00 0.000000 \n",
|
||
"2 0.00 0.000 0.00 0.000000 \n",
|
||
"3 20698.97 27252.124 177077.31 1362.133154 \n",
|
||
"4 0.00 0.000 0.00 0.000000 \n",
|
||
"\n",
|
||
" net_flow_qty_vol rel_intensity netflow_to_aum n_tx_total \\\n",
|
||
"0 0.000000 0.000000 0.000000 0 \n",
|
||
"1 0.000000 0.000000 0.000000 0 \n",
|
||
"2 0.000000 0.000000 0.000000 0 \n",
|
||
"3 3508.455222 0.056109 0.011473 161 \n",
|
||
"4 0.000000 0.000000 0.000000 0 \n",
|
||
"\n",
|
||
" log_aum_qty_mean log_gross_flow_qty_mean gross_flow_to_aum \n",
|
||
"0 0.000000 0.000000 0.00000 \n",
|
||
"1 0.000000 0.000000 0.00000 \n",
|
||
"2 0.000000 0.000000 0.00000 \n",
|
||
"3 9.979042 7.217541 8.20994 \n",
|
||
"4 0.000000 0.000000 0.00000 "
|
||
]
|
||
},
|
||
"execution_count": 13,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"eps = 1e-9 \n",
|
||
"\n",
|
||
"# 1) Active month indicator: did the client trade this month?\n",
|
||
"df_month[\"active_month\"] = (df_month[\"gross_flow_qty\"] > 0).astype(int)\n",
|
||
"\n",
|
||
"#client avec beaucoup de mois à 0 → “stable / dormant”\n",
|
||
"#client actif presque tous les mois → “rebalancer / institutionnel actif”\n",
|
||
"\n",
|
||
"\n",
|
||
"# 2) Monthly relative intensity (turnover proxy in quantity terms) : Mesurer l’intensité de trading relativement à la taille et pouvoir ocmparer client petit avec client plus gros\n",
|
||
"df_month[\"rel_intensity_m\"] = df_month[\"gross_flow_qty\"] / (df_month[\"aum_qty\"].abs() + eps)\n",
|
||
"\n",
|
||
"# 3) Monthly net flow ratio (directional change): sert a Capturer la direction de la dynamique\n",
|
||
"df_month[\"netflow_to_aum_m\"] = df_month[\"net_flow_qty\"] / (df_month[\"aum_qty\"].abs() + eps)\n",
|
||
"\n",
|
||
"# 4) Aggregate to client-level features (1 row per client)\n",
|
||
"df_client_feat = (\n",
|
||
" df_month.groupby(ID_COL, as_index=False)\n",
|
||
" .agg(\n",
|
||
" # Coverage / activity\n",
|
||
" n_months=(\"month\", \"nunique\"),\n",
|
||
" n_active_months=(\"active_month\", \"sum\"),\n",
|
||
" flow_freq=(\"active_month\", \"mean\"),\n",
|
||
"\n",
|
||
" # Size in quantity terms\n",
|
||
" aum_qty_mean=(\"aum_qty\", \"mean\"),\n",
|
||
" aum_qty_median=(\"aum_qty\", \"median\"),\n",
|
||
"\n",
|
||
" # Flows in quantity terms\n",
|
||
" net_flow_qty_sum=(\"net_flow_qty\", \"sum\"),\n",
|
||
" gross_flow_qty_sum=(\"gross_flow_qty\", \"sum\"),\n",
|
||
" gross_flow_qty_mean=(\"gross_flow_qty\", \"mean\"),\n",
|
||
"\n",
|
||
" # Dispersion / volatility proxy\n",
|
||
" net_flow_qty_vol=(\"net_flow_qty\", \"std\"),\n",
|
||
" rel_intensity=(\"rel_intensity_m\", \"mean\"),\n",
|
||
" netflow_to_aum=(\"netflow_to_aum_m\", \"mean\"),\n",
|
||
"\n",
|
||
" # Trading frequency proxy\n",
|
||
" n_tx_total=(\"n_tx\", \"sum\"),\n",
|
||
" )\n",
|
||
")\n",
|
||
"\n",
|
||
"# 5) Clean NaNs due to std on constant series\n",
|
||
"df_client_feat[\"net_flow_qty_vol\"] = df_client_feat[\"net_flow_qty_vol\"].fillna(0.0)\n",
|
||
"\n",
|
||
"# 6) Log transforms (useful because distributions are heavy-tailed)\n",
|
||
"df_client_feat[\"log_aum_qty_mean\"] = np.log1p(df_client_feat[\"aum_qty_mean\"].clip(lower=0))\n",
|
||
"df_client_feat[\"log_gross_flow_qty_mean\"] = np.log1p(df_client_feat[\"gross_flow_qty_mean\"].clip(lower=0))\n",
|
||
"\n",
|
||
"# 7) Global turnover proxy\n",
|
||
"df_client_feat[\"gross_flow_to_aum\"] = df_client_feat[\"gross_flow_qty_sum\"] / (df_client_feat[\"aum_qty_mean\"].abs() + eps)\n",
|
||
"\n",
|
||
"df_client_feat.head()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "7650f113-fb48-41c0-87a4-5b9509086ffd",
|
||
"metadata": {},
|
||
"source": [
|
||
"Pour faire du clustering, il faut un tableau où :\n",
|
||
"- 1 ligne = 1 “objet à classer” (ici : un client / compte)\n",
|
||
"- colonnes = variables comparables entre clients\n",
|
||
"\n",
|
||
"Ce code sert à :\n",
|
||
"1- transformer une série temporelle mensuelle (client × mois)\n",
|
||
"2- en un vecteur de caractéristiques (client → features)\n",
|
||
"! utilisable en clustering."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 14,
|
||
"id": "f07707c2-480c-4dd1-8767-95d3582b3149",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"(15491, 16)"
|
||
]
|
||
},
|
||
"execution_count": 14,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"df_client_feat.shape"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 15,
|
||
"id": "5e993274-eace-4527-b824-94144f155688",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Filtered clients: (9743, 16)\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"dfc = df_client_feat.copy()\n",
|
||
"\n",
|
||
"# Minimal filters (adjust if needed)\n",
|
||
"dfc = dfc[(dfc[\"n_months\"] >= 6)] # at least 6 observed months\n",
|
||
"dfc = dfc[(dfc[\"aum_qty_mean\"].abs() > 0)] # avoid zero holdings\n",
|
||
"print(\"Filtered clients:\", dfc.shape)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "7ff3ffbc-58d8-482f-a027-80b032897c33",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Baseline clustering 1"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 16,
|
||
"id": "1a9e1bcb-8f5d-4fce-8925-ef6bfea84967",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Q33: 0.05374559928443659 Q66: 25.599990882352966\n",
|
||
"seg_quantiles\n",
|
||
"High-flow 3313\n",
|
||
"Low-flow 3215\n",
|
||
"Intermediate-flow 3215\n",
|
||
"Name: count, dtype: int64\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Baseline 1 variable: average monthly gross traded quantity\n",
|
||
"x = dfc[\"gross_flow_qty_mean\"].copy()\n",
|
||
"\n",
|
||
"q33, q66 = x.quantile([0.33, 0.66])\n",
|
||
"\n",
|
||
"dfc[\"seg_quantiles\"] = pd.cut(\n",
|
||
" x,\n",
|
||
" bins=[-np.inf, q33, q66, np.inf],\n",
|
||
" labels=[\"Low-flow\", \"Intermediate-flow\", \"High-flow\"]\n",
|
||
")\n",
|
||
"\n",
|
||
"print(\"Q33:\", q33, \" Q66:\", q66)\n",
|
||
"print(dfc[\"seg_quantiles\"].value_counts(dropna=False))"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 17,
|
||
"id": "5236feba-7d3e-4224-9485-0449bf35c86f",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"plt.figure()\n",
|
||
"plt.hist(dfc[\"gross_flow_qty_mean\"], bins=100)\n",
|
||
"plt.axvline(q33, linestyle=\"--\")\n",
|
||
"plt.axvline(q66, linestyle=\"--\")\n",
|
||
"plt.xlabel(\"Average monthly gross flow (quantity)\")\n",
|
||
"plt.ylabel(\"Count\")\n",
|
||
"plt.title(\"Distribution of average monthly gross flows (quantity) with quantile thresholds\")\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"#X= activite moyenen mensuelle , Y = combien de client ont cette valeurde X"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 18,
|
||
"id": "4e6a7679-386a-468c-84f8-5179eab21f82",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"plt.figure()\n",
|
||
"plt.hist(np.log1p(dfc[\"gross_flow_qty_mean\"]), bins=100)\n",
|
||
"plt.axvline(np.log1p(q33), linestyle=\"--\")\n",
|
||
"plt.axvline(np.log1p(q66), linestyle=\"--\")\n",
|
||
"plt.xlabel(\"log(1 + avg monthly gross flow) (quantity)\")\n",
|
||
"plt.ylabel(\"Count\")\n",
|
||
"plt.title(\"Log-distribution of average monthly gross flows (quantity)\")\n",
|
||
"plt.show()\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 19,
|
||
"id": "658f167f-1e8d-4c34-ae7e-bd1d4759d5c4",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"(array([415., 187., 172., 173., 162., 136., 125., 126., 112., 136., 134.,\n",
|
||
" 153., 142., 150., 154., 158., 163., 174., 135., 133., 153., 128.,\n",
|
||
" 130., 139., 120., 116., 118., 101., 82., 89., 100., 90., 82.,\n",
|
||
" 104., 86., 104., 90., 82., 87., 70., 78., 85., 66., 62.,\n",
|
||
" 91., 74., 63., 78., 62., 70., 51., 65., 48., 44., 42.,\n",
|
||
" 53., 52., 43., 35., 35., 22., 27., 29., 25., 21., 21.,\n",
|
||
" 25., 15., 19., 9., 8., 7., 9., 7., 4., 7., 4.,\n",
|
||
" 1., 4., 2., 2., 1., 0., 5., 2., 2., 0., 2.,\n",
|
||
" 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
|
||
" 1.]),\n",
|
||
" array([1.24999219e-05, 1.51704407e-01, 3.03396315e-01, 4.55088222e-01,\n",
|
||
" 6.06780130e-01, 7.58472037e-01, 9.10163945e-01, 1.06185585e+00,\n",
|
||
" 1.21354776e+00, 1.36523967e+00, 1.51693157e+00, 1.66862348e+00,\n",
|
||
" 1.82031539e+00, 1.97200730e+00, 2.12369920e+00, 2.27539111e+00,\n",
|
||
" 2.42708302e+00, 2.57877493e+00, 2.73046683e+00, 2.88215874e+00,\n",
|
||
" 3.03385065e+00, 3.18554256e+00, 3.33723446e+00, 3.48892637e+00,\n",
|
||
" 3.64061828e+00, 3.79231019e+00, 3.94400209e+00, 4.09569400e+00,\n",
|
||
" 4.24738591e+00, 4.39907782e+00, 4.55076972e+00, 4.70246163e+00,\n",
|
||
" 4.85415354e+00, 5.00584545e+00, 5.15753735e+00, 5.30922926e+00,\n",
|
||
" 5.46092117e+00, 5.61261308e+00, 5.76430498e+00, 5.91599689e+00,\n",
|
||
" 6.06768880e+00, 6.21938071e+00, 6.37107261e+00, 6.52276452e+00,\n",
|
||
" 6.67445643e+00, 6.82614834e+00, 6.97784024e+00, 7.12953215e+00,\n",
|
||
" 7.28122406e+00, 7.43291597e+00, 7.58460787e+00, 7.73629978e+00,\n",
|
||
" 7.88799169e+00, 8.03968360e+00, 8.19137550e+00, 8.34306741e+00,\n",
|
||
" 8.49475932e+00, 8.64645123e+00, 8.79814313e+00, 8.94983504e+00,\n",
|
||
" 9.10152695e+00, 9.25321886e+00, 9.40491076e+00, 9.55660267e+00,\n",
|
||
" 9.70829458e+00, 9.85998649e+00, 1.00116784e+01, 1.01633703e+01,\n",
|
||
" 1.03150622e+01, 1.04667541e+01, 1.06184460e+01, 1.07701379e+01,\n",
|
||
" 1.09218298e+01, 1.10735217e+01, 1.12252137e+01, 1.13769056e+01,\n",
|
||
" 1.15285975e+01, 1.16802894e+01, 1.18319813e+01, 1.19836732e+01,\n",
|
||
" 1.21353651e+01, 1.22870570e+01, 1.24387489e+01, 1.25904408e+01,\n",
|
||
" 1.27421327e+01, 1.28938246e+01, 1.30455165e+01, 1.31972085e+01,\n",
|
||
" 1.33489004e+01, 1.35005923e+01, 1.36522842e+01, 1.38039761e+01,\n",
|
||
" 1.39556680e+01, 1.41073599e+01, 1.42590518e+01, 1.44107437e+01,\n",
|
||
" 1.45624356e+01, 1.47141275e+01, 1.48658194e+01, 1.50175113e+01,\n",
|
||
" 1.51692032e+01]),\n",
|
||
" <BarContainer object of 100 artists>)"
|
||
]
|
||
},
|
||
"execution_count": 19,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"plt.hist(np.log1p(dfc.loc[dfc[\"gross_flow_qty_mean\"]>0, \"gross_flow_qty_mean\"]), bins=100)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 20,
|
||
"id": "42eb180e-a512-4d5c-8179-91b88e3067df",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/tmp/ipykernel_8776/4281614443.py:3: MatplotlibDeprecationWarning: The 'labels' parameter of boxplot() has been renamed 'tick_labels' since Matplotlib 3.9; support for the old name will be dropped in 3.11.\n",
|
||
" plt.boxplot(\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"' Ce que tu vois sur ton graphe\\n\\nLes trois segments contiennent des outliers énormes.\\nIl y a une forte dispersion dans chaque segment.\\nMême dans Low-flow et Mid-flow, tu as des comptes avec AUM très grandes (points très haut).\\nEt inversement, dans High-flow, tu as aussi des comptes relativement petits.\\n\\nC’est ça qui justifie la phrase :\\n“Flow intensity alone does not uniquely define client type because accounts of very different sizes may fall in the same flow segment.”\\n'"
|
||
]
|
||
},
|
||
"execution_count": 20,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"#limite client size \n",
|
||
"plt.figure()\n",
|
||
"plt.boxplot(\n",
|
||
" [dfc.loc[dfc[\"seg_quantiles\"]==s, \"aum_qty_mean\"].dropna()\n",
|
||
" for s in [\"Low-flow\",\"Intermediate-flow\",\"High-flow\"]],\n",
|
||
" labels=[\"Low\",\"Mid\",\"High\"]\n",
|
||
")\n",
|
||
"plt.yscale(\"log\")\n",
|
||
"plt.ylabel(\"Mean AUM (quantity) [log scale]\")\n",
|
||
"plt.title(\"Client size differs widely within flow-intensity segments (quantity AUM)\")\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"\n",
|
||
"''' Ce que tu vois sur ton graphe\n",
|
||
"\n",
|
||
"Les trois segments contiennent des outliers énormes.\n",
|
||
"Il y a une forte dispersion dans chaque segment.\n",
|
||
"Même dans Low-flow et Mid-flow, tu as des comptes avec AUM très grandes (points très haut).\n",
|
||
"Et inversement, dans High-flow, tu as aussi des comptes relativement petits.\n",
|
||
"\n",
|
||
"C’est ça qui justifie la phrase :\n",
|
||
"“Flow intensity alone does not uniquely define client type because accounts of very different sizes may fall in the same flow segment.”\n",
|
||
"'''"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ae004a9f-3a5b-4570-a07c-2f4882fff3c0",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Baseline 2"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 21,
|
||
"id": "0e10f5a2-d678-484e-8789-a907635a5a20",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"count 9743.000000\n",
|
||
"mean 59.264703\n",
|
||
"std 38.537320\n",
|
||
"min 6.000000\n",
|
||
"25% 26.000000\n",
|
||
"50% 50.000000\n",
|
||
"75% 80.000000\n",
|
||
"max 130.000000\n",
|
||
"Name: n_months, dtype: float64"
|
||
]
|
||
},
|
||
"execution_count": 21,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"dfc[\"n_months\"].describe()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 22,
|
||
"id": "0eb64794-a3b2-4aa3-a43c-62a39eab35e2",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"seg_2D\n",
|
||
"Highly active (high int, high freq) 3624\n",
|
||
"Dormant (low int, low freq) 3588\n",
|
||
"Small rebalancers (low int, high freq) 1283\n",
|
||
"Occasional large movers (high int, low freq) 1248\n",
|
||
"Name: count, dtype: int64\n",
|
||
"thr_int: 1.7333333333210905 thr_freq: 0.041666666666666664\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"dfc[\"rel_intensity_total\"] = dfc[\"gross_flow_to_aum\"] # turnover proxy\n",
|
||
"dfc[\"frequency\"] = dfc[\"flow_freq\"] # share of active months\n",
|
||
"\n",
|
||
"# Thresholds: medians (simple + explainable)\n",
|
||
"thr_int = dfc[\"rel_intensity_total\"].median()\n",
|
||
"thr_freq = dfc[\"frequency\"].median()\n",
|
||
"\n",
|
||
"def quadrant(row):\n",
|
||
" low_int = row[\"rel_intensity_total\"] < thr_int\n",
|
||
" low_frq = row[\"frequency\"] < thr_freq\n",
|
||
"\n",
|
||
" if low_int and low_frq:\n",
|
||
" return \"Dormant (low int, low freq)\"\n",
|
||
" if low_int and (not low_frq):\n",
|
||
" return \"Small rebalancers (low int, high freq)\"\n",
|
||
" if (not low_int) and low_frq:\n",
|
||
" return \"Occasional large movers (high int, low freq)\"\n",
|
||
" return \"Highly active (high int, high freq)\"\n",
|
||
"\n",
|
||
"dfc[\"seg_2D\"] = dfc.apply(quadrant, axis=1)\n",
|
||
"\n",
|
||
"print(dfc[\"seg_2D\"].value_counts())\n",
|
||
"print(\"thr_int:\", thr_int, \"thr_freq:\", thr_freq)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 26,
|
||
"id": "2850c428-c9bf-42fa-87d1-fd8d5d10b016",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"plt.figure()\n",
|
||
"for name, g in dfc.groupby(\"seg_2D\"):\n",
|
||
" plt.scatter(g[\"frequency\"], g[\"rel_intensity_total\"], s=10, label=name)\n",
|
||
"\n",
|
||
"plt.yscale(\"log\")\n",
|
||
"plt.axvline(thr_freq, linestyle=\"--\")\n",
|
||
"plt.axhline(thr_int, linestyle=\"--\")\n",
|
||
"plt.xlabel(\"Activity frequency (share of active months)\")\n",
|
||
"plt.ylabel(\"Gross flow / mean AUM (quantity) [log scale]\")\n",
|
||
"plt.title(\"2D behavioral segmentation: relative intensity vs frequency\")\n",
|
||
"plt.legend(markerscale=2)\n",
|
||
"plt.show()\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 27,
|
||
"id": "d05ac724-6aab-4ea9-bce5-d1b0252148a5",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>n_clients</th>\n",
|
||
" <th>aum_qty_med</th>\n",
|
||
" <th>gross_flow_qty_med</th>\n",
|
||
" <th>freq_med</th>\n",
|
||
" <th>rel_int_med</th>\n",
|
||
" <th>n_tx_med</th>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>seg_2D</th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>Highly active (high int, high freq)</th>\n",
|
||
" <td>3624</td>\n",
|
||
" <td>823.825696</td>\n",
|
||
" <td>105.945544</td>\n",
|
||
" <td>0.157379</td>\n",
|
||
" <td>5.991354</td>\n",
|
||
" <td>17.0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>Dormant (low int, low freq)</th>\n",
|
||
" <td>3588</td>\n",
|
||
" <td>102.161465</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>Small rebalancers (low int, high freq)</th>\n",
|
||
" <td>1283</td>\n",
|
||
" <td>543.424242</td>\n",
|
||
" <td>12.149239</td>\n",
|
||
" <td>0.100000</td>\n",
|
||
" <td>0.911442</td>\n",
|
||
" <td>15.0</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>Occasional large movers (high int, low freq)</th>\n",
|
||
" <td>1248</td>\n",
|
||
" <td>39.278192</td>\n",
|
||
" <td>4.349875</td>\n",
|
||
" <td>0.025000</td>\n",
|
||
" <td>4.497671</td>\n",
|
||
" <td>1.0</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" n_clients aum_qty_med \\\n",
|
||
"seg_2D \n",
|
||
"Highly active (high int, high freq) 3624 823.825696 \n",
|
||
"Dormant (low int, low freq) 3588 102.161465 \n",
|
||
"Small rebalancers (low int, high freq) 1283 543.424242 \n",
|
||
"Occasional large movers (high int, low freq) 1248 39.278192 \n",
|
||
"\n",
|
||
" gross_flow_qty_med freq_med \\\n",
|
||
"seg_2D \n",
|
||
"Highly active (high int, high freq) 105.945544 0.157379 \n",
|
||
"Dormant (low int, low freq) 0.000000 0.000000 \n",
|
||
"Small rebalancers (low int, high freq) 12.149239 0.100000 \n",
|
||
"Occasional large movers (high int, low freq) 4.349875 0.025000 \n",
|
||
"\n",
|
||
" rel_int_med n_tx_med \n",
|
||
"seg_2D \n",
|
||
"Highly active (high int, high freq) 5.991354 17.0 \n",
|
||
"Dormant (low int, low freq) 0.000000 0.0 \n",
|
||
"Small rebalancers (low int, high freq) 0.911442 15.0 \n",
|
||
"Occasional large movers (high int, low freq) 4.497671 1.0 "
|
||
]
|
||
},
|
||
"execution_count": 27,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"profile_2d = (\n",
|
||
" dfc.groupby(\"seg_2D\")\n",
|
||
" .agg(\n",
|
||
" n_clients=(ID_COL, \"count\"),\n",
|
||
" aum_qty_med=(\"aum_qty_mean\",\"median\"),\n",
|
||
" gross_flow_qty_med=(\"gross_flow_qty_mean\",\"median\"),\n",
|
||
" freq_med=(\"frequency\",\"median\"),\n",
|
||
" rel_int_med=(\"rel_intensity_total\",\"median\"),\n",
|
||
" n_tx_med=(\"n_tx_total\",\"median\"),\n",
|
||
" )\n",
|
||
" .sort_values(\"n_clients\", ascending=False)\n",
|
||
")\n",
|
||
"profile_2d\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "4ceeeec6-4f52-4b8b-9c14-2419181d0a5f",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Clustering 2 :KMEANS"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 23,
|
||
"id": "6afd24fe-0886-445c-a330-6a537d6aaa1a",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Clustering matrix shape: (9743, 6)\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"import numpy as np\n",
|
||
"import pandas as pd\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"\n",
|
||
"from sklearn.preprocessing import StandardScaler\n",
|
||
"from sklearn.cluster import KMeans\n",
|
||
"from sklearn.metrics import silhouette_score\n",
|
||
"from sklearn.mixture import GaussianMixture\n",
|
||
"\n",
|
||
"# Safety: ensure baseline-2 columns exist\n",
|
||
"dfc = dfc.copy()\n",
|
||
"dfc[\"frequency\"] = dfc[\"flow_freq\"]\n",
|
||
"dfc[\"rel_intensity_total\"] = dfc[\"gross_flow_to_aum\"]\n",
|
||
"\n",
|
||
"# Choose a compact, interpretable feature set (quantity-based)\n",
|
||
"features = [\n",
|
||
" \"log_aum_qty_mean\", # size (log)\n",
|
||
" \"log_gross_flow_qty_mean\", # activity intensity (log)\n",
|
||
" \"frequency\", # activity frequency\n",
|
||
" \"rel_intensity_total\", # turnover proxy\n",
|
||
" \"net_flow_qty_vol\", # volatility of net flows\n",
|
||
" \"n_tx_total\", # total number of transactions\n",
|
||
"]\n",
|
||
"\n",
|
||
"# Build X (drop NaNs/Infs)\n",
|
||
"X = (dfc[features]\n",
|
||
" .replace([np.inf, -np.inf], np.nan)\n",
|
||
" .dropna()\n",
|
||
" .copy())\n",
|
||
"\n",
|
||
"# Keep IDs aligned\n",
|
||
"ids = dfc.loc[X.index, ID_COL].copy()\n",
|
||
"\n",
|
||
"# Standardize (critical for distance-based clustering)\n",
|
||
"scaler = StandardScaler()\n",
|
||
"X_scaled = scaler.fit_transform(X)\n",
|
||
"\n",
|
||
"print(\"Clustering matrix shape:\", X_scaled.shape)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 24,
|
||
"id": "d2e58d79-0638-4cfa-9149-f3bfe5f89399",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Best K by silhouette: 3\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"k_range = range(2, 11)\n",
|
||
"inertias = []\n",
|
||
"silhouettes = []\n",
|
||
"\n",
|
||
"for k in k_range:\n",
|
||
" km = KMeans(n_clusters=k, n_init=30, random_state=42)\n",
|
||
" labels = km.fit_predict(X_scaled)\n",
|
||
" inertias.append(km.inertia_)\n",
|
||
" silhouettes.append(silhouette_score(X_scaled, labels))\n",
|
||
"\n",
|
||
"# Elbow plot\n",
|
||
"plt.figure()\n",
|
||
"plt.plot(list(k_range), inertias, marker=\"o\")\n",
|
||
"plt.xlabel(\"Number of clusters K\")\n",
|
||
"plt.ylabel(\"Inertia (within-cluster SSE)\")\n",
|
||
"plt.title(\"Elbow curve for K-means\")\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"# Silhouette plot\n",
|
||
"plt.figure()\n",
|
||
"plt.plot(list(k_range), silhouettes, marker=\"o\")\n",
|
||
"plt.xlabel(\"Number of clusters K\")\n",
|
||
"plt.ylabel(\"Silhouette score\")\n",
|
||
"plt.title(\"Silhouette score for K-means\")\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"best_k = list(k_range)[int(np.argmax(silhouettes))]\n",
|
||
"print(\"Best K by silhouette:\", best_k)\n",
|
||
"\n",
|
||
"\n",
|
||
"''' Ce que c’est :\n",
|
||
"Inertia = somme des distances intra-cluster (SSE).\n",
|
||
"Plus elle baisse, plus les clusters sont “serrés”.\n",
|
||
"\n",
|
||
"Comment lire :\n",
|
||
"Quand K augmente, inertia baisse toujours (normal).\n",
|
||
"On cherche un “coude” : à partir d’un certain K, ajouter des clusters apporte peu\n",
|
||
"'''"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ddfb03e9-a9ff-463d-b653-f752773aacd9",
|
||
"metadata": {},
|
||
"source": [
|
||
"Valeurs Silhouette:\n",
|
||
"proche de 1 : très bon\n",
|
||
"0 : clusters se chevauchent\n",
|
||
"négatif : mauvais clustering"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 25,
|
||
"id": "ea340dca-a76a-4d50-8973-af1dca34c376",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>n_clients</th>\n",
|
||
" <th>aum_qty_med</th>\n",
|
||
" <th>freq_med</th>\n",
|
||
" <th>rel_int_med</th>\n",
|
||
" <th>gross_flow_med</th>\n",
|
||
" <th>n_tx_med</th>\n",
|
||
" <th>vol_med</th>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>cluster_kmeans</th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>0.0</th>\n",
|
||
" <td>7718</td>\n",
|
||
" <td>104.983133</td>\n",
|
||
" <td>0.024390</td>\n",
|
||
" <td>1.039412e+00</td>\n",
|
||
" <td>0.774805</td>\n",
|
||
" <td>1.0</td>\n",
|
||
" <td>5.080794</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1.0</th>\n",
|
||
" <td>2024</td>\n",
|
||
" <td>14064.797371</td>\n",
|
||
" <td>0.575379</td>\n",
|
||
" <td>4.591024e+00</td>\n",
|
||
" <td>1100.212569</td>\n",
|
||
" <td>198.5</td>\n",
|
||
" <td>2291.806790</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2.0</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>0.000079</td>\n",
|
||
" <td>0.014706</td>\n",
|
||
" <td>1.257232e+06</td>\n",
|
||
" <td>1.468240</td>\n",
|
||
" <td>1.0</td>\n",
|
||
" <td>12.107415</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" n_clients aum_qty_med freq_med rel_int_med \\\n",
|
||
"cluster_kmeans \n",
|
||
"0.0 7718 104.983133 0.024390 1.039412e+00 \n",
|
||
"1.0 2024 14064.797371 0.575379 4.591024e+00 \n",
|
||
"2.0 1 0.000079 0.014706 1.257232e+06 \n",
|
||
"\n",
|
||
" gross_flow_med n_tx_med vol_med \n",
|
||
"cluster_kmeans \n",
|
||
"0.0 0.774805 1.0 5.080794 \n",
|
||
"1.0 1100.212569 198.5 2291.806790 \n",
|
||
"2.0 1.468240 1.0 12.107415 "
|
||
]
|
||
},
|
||
"execution_count": 25,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"km = KMeans(n_clusters=best_k, n_init=50, random_state=42)\n",
|
||
"labels_km = km.fit_predict(X_scaled)\n",
|
||
"\n",
|
||
"dfc.loc[X.index, \"cluster_kmeans\"] = labels_km\n",
|
||
"\n",
|
||
"# Profiling table (medians = robust to outliers)\n",
|
||
"k_profile = (\n",
|
||
" dfc.loc[X.index]\n",
|
||
" .groupby(\"cluster_kmeans\")\n",
|
||
" .agg(\n",
|
||
" n_clients=(ID_COL, \"count\"),\n",
|
||
" aum_qty_med=(\"aum_qty_mean\", \"median\"),\n",
|
||
" freq_med=(\"frequency\", \"median\"),\n",
|
||
" rel_int_med=(\"rel_intensity_total\", \"median\"),\n",
|
||
" gross_flow_med=(\"gross_flow_qty_mean\", \"median\"),\n",
|
||
" n_tx_med=(\"n_tx_total\", \"median\"),\n",
|
||
" vol_med=(\"net_flow_qty_vol\", \"median\"),\n",
|
||
" )\n",
|
||
" .sort_values(\"n_clients\", ascending=False)\n",
|
||
")\n",
|
||
"\n",
|
||
"k_profile\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "7702b855-8c02-486e-9c28-0e8c1b872a73",
|
||
"metadata": {},
|
||
"source": [
|
||
"## PARTIE 3 — GMM (Gaussian Mixture Model)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "810e73ed-f479-4f76-806c-3c18d281a931",
|
||
"metadata": {},
|
||
"source": [
|
||
"BIC/AIC = critères pour choisir le nombre de composantes (clusters) dans un modèle probabiliste.\n",
|
||
"Plus bas = meilleur compromis “fit” vs “complexité”."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 26,
|
||
"id": "ba9df953-cc88-4907-9fe7-b88978f318b3",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Best K by BIC: 10\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"bics, aics = [], []\n",
|
||
"models = {}\n",
|
||
"\n",
|
||
"for k in k_range:\n",
|
||
" gmm = GaussianMixture(n_components=k, covariance_type=\"full\", random_state=42)\n",
|
||
" gmm.fit(X_scaled)\n",
|
||
" bics.append(gmm.bic(X_scaled))\n",
|
||
" aics.append(gmm.aic(X_scaled))\n",
|
||
" models[k] = gmm\n",
|
||
"\n",
|
||
"plt.figure()\n",
|
||
"plt.plot(list(k_range), bics, marker=\"o\", label=\"BIC\")\n",
|
||
"plt.plot(list(k_range), aics, marker=\"o\", label=\"AIC\")\n",
|
||
"plt.xlabel(\"Number of components K\")\n",
|
||
"plt.ylabel(\"Criterion value\")\n",
|
||
"plt.title(\"GMM model selection (lower is better)\")\n",
|
||
"plt.legend()\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"best_k_gmm = list(k_range)[int(np.argmin(bics))]\n",
|
||
"print(\"Best K by BIC:\", best_k_gmm)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ce9b76de-2191-4528-8913-c2bb1911e97c",
|
||
"metadata": {},
|
||
"source": [
|
||
"Graphe :\n",
|
||
"- BIC/AIC continuent de baisser jusqu’à K=10. ==> Donc “Best K by BIC = 10”.\n",
|
||
"\n",
|
||
"Attention importante :\n",
|
||
"Quand BIC baisse toujours avec K, ça peut signifier :\n",
|
||
"- les données ont beaucoup de structure fine / multi-groupes\n",
|
||
"- OU le modèle “découpe” surtout des outliers / sous-groupes très rares\n",
|
||
"- OU le modèle est en train de sur-segmenter des extrêmes (typique finance)\n",
|
||
"Donc K=10 n’est pas automatiquement “mieux” au sens business, c’est “mieux” pour le fit probabiliste.\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 27,
|
||
"id": "47e0a0bd-0a26-4e56-91cb-5780a3c1fb86",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Share of ambiguous assignments (max prob < 0.6): 1.10%\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"gmm = models[best_k_gmm]\n",
|
||
"labels_gmm = gmm.predict(X_scaled)\n",
|
||
"proba = gmm.predict_proba(X_scaled)\n",
|
||
"max_proba = proba.max(axis=1)\n",
|
||
"\n",
|
||
"dfc.loc[X.index, \"cluster_gmm\"] = labels_gmm\n",
|
||
"dfc.loc[X.index, \"cluster_confidence\"] = max_proba\n",
|
||
"\n",
|
||
"# Confidence histogram\n",
|
||
"plt.figure()\n",
|
||
"plt.hist(max_proba, bins=50)\n",
|
||
"plt.xlabel(\"Max posterior probability\")\n",
|
||
"plt.ylabel(\"Count\")\n",
|
||
"plt.title(\"GMM assignment confidence\")\n",
|
||
"plt.show()\n",
|
||
"\n",
|
||
"ambiguous_share = (max_proba < 0.6).mean()\n",
|
||
"print(f\"Share of ambiguous assignments (max prob < 0.6): {ambiguous_share:.2%}\")\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 28,
|
||
"id": "d4254585-959f-4f79-9c08-73e6d7be25c5",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"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>n_clients</th>\n",
|
||
" <th>aum_qty_med</th>\n",
|
||
" <th>freq_med</th>\n",
|
||
" <th>rel_int_med</th>\n",
|
||
" <th>gross_flow_med</th>\n",
|
||
" <th>n_tx_med</th>\n",
|
||
" <th>vol_med</th>\n",
|
||
" <th>conf_med</th>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>cluster_gmm</th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" <th></th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>9.0</th>\n",
|
||
" <td>3509</td>\n",
|
||
" <td>7.949166e+01</td>\n",
|
||
" <td>0.041667</td>\n",
|
||
" <td>3.071429e+00</td>\n",
|
||
" <td>6.687053e+00</td>\n",
|
||
" <td>2.0</td>\n",
|
||
" <td>3.277044e+01</td>\n",
|
||
" <td>0.998364</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>1.0</th>\n",
|
||
" <td>2953</td>\n",
|
||
" <td>9.121500e+01</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>0.0</td>\n",
|
||
" <td>0.000000e+00</td>\n",
|
||
" <td>0.999999</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>6.0</th>\n",
|
||
" <td>1426</td>\n",
|
||
" <td>5.339202e+02</td>\n",
|
||
" <td>0.224745</td>\n",
|
||
" <td>2.483775e+00</td>\n",
|
||
" <td>2.762873e+01</td>\n",
|
||
" <td>38.0</td>\n",
|
||
" <td>7.746854e+01</td>\n",
|
||
" <td>0.999916</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>5.0</th>\n",
|
||
" <td>787</td>\n",
|
||
" <td>1.454968e+04</td>\n",
|
||
" <td>0.673267</td>\n",
|
||
" <td>5.310471e+00</td>\n",
|
||
" <td>1.064410e+03</td>\n",
|
||
" <td>485.0</td>\n",
|
||
" <td>1.867233e+03</td>\n",
|
||
" <td>0.999937</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>0.0</th>\n",
|
||
" <td>568</td>\n",
|
||
" <td>5.309168e+03</td>\n",
|
||
" <td>0.087500</td>\n",
|
||
" <td>6.195235e+00</td>\n",
|
||
" <td>6.734335e+02</td>\n",
|
||
" <td>5.0</td>\n",
|
||
" <td>2.910116e+03</td>\n",
|
||
" <td>0.998759</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>8.0</th>\n",
|
||
" <td>223</td>\n",
|
||
" <td>1.840341e+05</td>\n",
|
||
" <td>0.680851</td>\n",
|
||
" <td>6.853229e+00</td>\n",
|
||
" <td>1.521262e+04</td>\n",
|
||
" <td>418.0</td>\n",
|
||
" <td>3.652430e+04</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>4.0</th>\n",
|
||
" <td>198</td>\n",
|
||
" <td>1.656004e+05</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>4.315459e+00</td>\n",
|
||
" <td>8.092373e+03</td>\n",
|
||
" <td>3570.0</td>\n",
|
||
" <td>9.626152e+03</td>\n",
|
||
" <td>0.999323</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>7.0</th>\n",
|
||
" <td>77</td>\n",
|
||
" <td>1.109350e+00</td>\n",
|
||
" <td>0.130435</td>\n",
|
||
" <td>1.462797e+02</td>\n",
|
||
" <td>9.281811e+01</td>\n",
|
||
" <td>15.0</td>\n",
|
||
" <td>2.553362e+02</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>2.0</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>7.941176e-05</td>\n",
|
||
" <td>0.014706</td>\n",
|
||
" <td>1.257232e+06</td>\n",
|
||
" <td>1.468240e+00</td>\n",
|
||
" <td>1.0</td>\n",
|
||
" <td>1.210741e+01</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>3.0</th>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1.319632e+08</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>3.814096e+00</td>\n",
|
||
" <td>3.871695e+06</td>\n",
|
||
" <td>27668.0</td>\n",
|
||
" <td>1.194728e+07</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
],
|
||
"text/plain": [
|
||
" n_clients aum_qty_med freq_med rel_int_med gross_flow_med \\\n",
|
||
"cluster_gmm \n",
|
||
"9.0 3509 7.949166e+01 0.041667 3.071429e+00 6.687053e+00 \n",
|
||
"1.0 2953 9.121500e+01 0.000000 0.000000e+00 0.000000e+00 \n",
|
||
"6.0 1426 5.339202e+02 0.224745 2.483775e+00 2.762873e+01 \n",
|
||
"5.0 787 1.454968e+04 0.673267 5.310471e+00 1.064410e+03 \n",
|
||
"0.0 568 5.309168e+03 0.087500 6.195235e+00 6.734335e+02 \n",
|
||
"8.0 223 1.840341e+05 0.680851 6.853229e+00 1.521262e+04 \n",
|
||
"4.0 198 1.656004e+05 1.000000 4.315459e+00 8.092373e+03 \n",
|
||
"7.0 77 1.109350e+00 0.130435 1.462797e+02 9.281811e+01 \n",
|
||
"2.0 1 7.941176e-05 0.014706 1.257232e+06 1.468240e+00 \n",
|
||
"3.0 1 1.319632e+08 1.000000 3.814096e+00 3.871695e+06 \n",
|
||
"\n",
|
||
" n_tx_med vol_med conf_med \n",
|
||
"cluster_gmm \n",
|
||
"9.0 2.0 3.277044e+01 0.998364 \n",
|
||
"1.0 0.0 0.000000e+00 0.999999 \n",
|
||
"6.0 38.0 7.746854e+01 0.999916 \n",
|
||
"5.0 485.0 1.867233e+03 0.999937 \n",
|
||
"0.0 5.0 2.910116e+03 0.998759 \n",
|
||
"8.0 418.0 3.652430e+04 1.000000 \n",
|
||
"4.0 3570.0 9.626152e+03 0.999323 \n",
|
||
"7.0 15.0 2.553362e+02 1.000000 \n",
|
||
"2.0 1.0 1.210741e+01 1.000000 \n",
|
||
"3.0 27668.0 1.194728e+07 1.000000 "
|
||
]
|
||
},
|
||
"execution_count": 28,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"g_profile = (\n",
|
||
" dfc.loc[X.index]\n",
|
||
" .groupby(\"cluster_gmm\")\n",
|
||
" .agg(\n",
|
||
" n_clients=(ID_COL, \"count\"),\n",
|
||
" aum_qty_med=(\"aum_qty_mean\", \"median\"),\n",
|
||
" freq_med=(\"frequency\", \"median\"),\n",
|
||
" rel_int_med=(\"rel_intensity_total\", \"median\"),\n",
|
||
" gross_flow_med=(\"gross_flow_qty_mean\", \"median\"),\n",
|
||
" n_tx_med=(\"n_tx_total\", \"median\"),\n",
|
||
" vol_med=(\"net_flow_qty_vol\", \"median\"),\n",
|
||
" conf_med=(\"cluster_confidence\", \"median\"),\n",
|
||
" )\n",
|
||
" .sort_values(\"n_clients\", ascending=False)\n",
|
||
")\n",
|
||
"\n",
|
||
"g_profile\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 29,
|
||
"id": "84d1790a-7901-473e-b367-b8464886ada2",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Silhouette K-means: 0.5887519415047723\n",
|
||
"Silhouette GMM: 0.138325887586186\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"#comapraison kmeans et gmm\n",
|
||
"\n",
|
||
"from sklearn.metrics import silhouette_score\n",
|
||
"\n",
|
||
"sil_km = silhouette_score(X_scaled, dfc.loc[X.index, \"cluster_kmeans\"])\n",
|
||
"sil_gmm = silhouette_score(X_scaled, dfc.loc[X.index, \"cluster_gmm\"])\n",
|
||
"\n",
|
||
"print(\"Silhouette K-means:\", sil_km)\n",
|
||
"print(\"Silhouette GMM:\", sil_gmm)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 30,
|
||
"id": "c4576042-09e9-4f03-b99b-883d8fa3426f",
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"DB index K-means (lower better): 0.5656170003935967\n",
|
||
"DB index GMM (lower better): 1.4187135041840173\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"''' \n",
|
||
"Davies–Bouldin index (plus bas = mieux)\n",
|
||
"Mesure similarité entre clusters.\n",
|
||
"Plus bas = clusters plus distincts.'''\n",
|
||
"\n",
|
||
"from sklearn.metrics import davies_bouldin_score\n",
|
||
"\n",
|
||
"db_km = davies_bouldin_score(X_scaled, dfc.loc[X.index, \"cluster_kmeans\"])\n",
|
||
"db_gmm = davies_bouldin_score(X_scaled, dfc.loc[X.index, \"cluster_gmm\"])\n",
|
||
"\n",
|
||
"print(\"DB index K-means (lower better):\", db_km)\n",
|
||
"print(\"DB index GMM (lower better):\", db_gmm)\n"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"id": "7d4d8264-359f-40c6-bb4d-4e26d1d74621",
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": []
|
||
}
|
||
],
|
||
"metadata": {
|
||
"kernelspec": {
|
||
"display_name": "Python 3 (ipykernel)",
|
||
"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": "3.13.11"
|
||
}
|
||
},
|
||
"nbformat": 4,
|
||
"nbformat_minor": 5
|
||
}
|