From 636c1acb8d24a2532f5e73ed75ca46c9ea7efd9c Mon Sep 17 00:00:00 2001 From: lmoraine-ensae Date: Thu, 5 Mar 2026 19:01:23 +0000 Subject: [PATCH] defi cleaning --- .../Challenge_cleaning-checkpoint.ipynb | 634 ++++ Challenge_cleaning.ipynb | 3110 +++++++++++++++++ Essai.ipynb | 2 +- Migration.ipynb | 2 +- rupture.ipynb | 2 +- 5 files changed, 3747 insertions(+), 3 deletions(-) create mode 100644 .ipynb_checkpoints/Challenge_cleaning-checkpoint.ipynb create mode 100644 Challenge_cleaning.ipynb diff --git a/.ipynb_checkpoints/Challenge_cleaning-checkpoint.ipynb b/.ipynb_checkpoints/Challenge_cleaning-checkpoint.ipynb new file mode 100644 index 0000000..700df89 --- /dev/null +++ b/.ipynb_checkpoints/Challenge_cleaning-checkpoint.ipynb @@ -0,0 +1,634 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "id": "4287b380-5359-49c7-ab95-ad346a3ad17a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Fichiers Flows : ['projet-bdc-data/carmignac/Flows ENSAE V1 -20251027.csv', 'projet-bdc-data/carmignac/Flows ENSAE V2 -20251105.csv']\n", + "Fichiers AUM : ['projet-bdc-data/carmignac/AUM ENSAE V1 -20251027.csv', 'projet-bdc-data/carmignac/AUM ENSAE V2 -20251105.csv']\n" + ] + } + ], + "source": [ + "# Import des données\n", + "\n", + "import os\n", + "import s3fs\n", + "import pandas as pd\n", + "\n", + "s3_ENDPOINT_URL = \"https://\" + os.environ[\"AWS_S3_ENDPOINT\"]\n", + "\n", + "fs = s3fs.S3FileSystem(client_kwargs={'endpoint_url': s3_ENDPOINT_URL})\n", + "\n", + "BUCKET = \"projet-bdc-data\"\n", + "carmignac_path = \"projet-bdc-data/carmignac\"\n", + "\n", + "# Liste des fichiers FLOWS\n", + "all_files = fs.ls(carmignac_path)\n", + "flows_files = [f for f in all_files if \"Flows\" in f and f.endswith(\".csv\")]\n", + "print(\"Fichiers Flows :\", flows_files)\n", + "\n", + "# Lire tous les fichiers dans un dictionnaire\n", + "flows_data = {}\n", + "for file_path in flows_files:\n", + " with fs.open(file_path, 'r') as f:\n", + " df = pd.read_csv(f, sep=';',low_memory=False)\n", + " flows_data[os.path.basename(file_path)] = df\n", + "\n", + "\n", + "# Liste des fichiers AUM\n", + "all_files = fs.ls(carmignac_path)\n", + "aum_files = [f for f in all_files if \"AUM\" in f and f.endswith(\".csv\")]\n", + "print(\"Fichiers AUM :\", aum_files)\n", + "\n", + "# Lire tous les fichiers dans un dictionnaire\n", + "aum_data = {}\n", + "for file_path in aum_files:\n", + " with fs.open(file_path, 'r') as f:\n", + " df = pd.read_csv(f, sep=';',low_memory=False)\n", + " aum_data[os.path.basename(file_path)] = df\n", + "\n", + "df = aum_data['AUM ENSAE V2 -20251105.csv']\n", + "dg = flows_data['Flows ENSAE V2 -20251105.csv']" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "0b66aee0-b726-4a57-9461-6a4550a625a8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Agreement - CodeCompany - IdCompany - Ultimate Parent IdRegistrar Account - IDRegistrar Account - RegionRegistrarAccount - CountryProduct - Asset TypeProduct - StrategyProduct - Legal StatusProduct - Is Dedie ?Product - FundProduct - Shareclass TypeProduct - Shareclass CurrencyProduct - IsinCentralisation DateQuantity - AUMValue - AUM CCYValue - AUM €
0003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-03-3135.36824648.666624648.6666
1003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-11-3035.36822413.055322413.0553
2003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-12-3135.36822051.240622051.2406
3003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032016-03-3135.36821626.117321626.1173
4003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032016-11-3035.36822489.450222489.4502
\n", + "
" + ], + "text/plain": [ + " Agreement - Code Company - Id Company - Ultimate Parent Id \\\n", + "0 003 166 166 \n", + "1 003 166 166 \n", + "2 003 166 166 \n", + "3 003 166 166 \n", + "4 003 166 166 \n", + "\n", + " Registrar Account - ID Registrar Account - Region \\\n", + "0 200000647 France \n", + "1 200000647 France \n", + "2 200000647 France \n", + "3 200000647 France \n", + "4 200000647 France \n", + "\n", + " RegistrarAccount - Country Product - Asset Type Product - Strategy \\\n", + "0 France Diversified Patrimoine \n", + "1 France Diversified Patrimoine \n", + "2 France Diversified Patrimoine \n", + "3 France Diversified Patrimoine \n", + "4 France Diversified Patrimoine \n", + "\n", + " Product - Legal Status Product - Is Dedie ? Product - Fund \\\n", + "0 FCP NO Carmignac Patrimoine \n", + "1 FCP NO Carmignac Patrimoine \n", + "2 FCP NO Carmignac Patrimoine \n", + "3 FCP NO Carmignac Patrimoine \n", + "4 FCP 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", + "\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 " + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "a5509e6e-ff10-4388-9fee-5cd49d01b60a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Agreement - CodeCompany - IdCompany - Ultimate Parent IdRegistrar Account - IDRegistrar Account - RegionRegistrarAccount - CountryProduct - Asset TypeProduct - StrategyProduct - Legal StatusProduct - Is Dedie ?...Centralisation DateQuantity - SubscriptionQuantity - RedemptionQuantity - NetFlowsValue Ccy - SubscriptionValue Ccy - RedemptionValue Ccy - NetFlowsValue € - SubscriptionValue € - RedemptionValue € - NetFlows
0003166166200127202FranceFranceEquityInvestissementSICAVNO...2020-11-051636.000.0001636.000280983.000.00280983.00280983.000.00280983.00
1003166166406533FranceFranceDiversifiedPatrimoineFCPNO...2015-03-09144.690.000144.69099985.130.0099985.1399985.130.0099985.13
2003166166406533FranceFranceEquityInvestissementFCPNO...2016-10-260.00-8.321-8.3210.00-9384.76-9384.760.00-9384.76-9384.76
3003166166406533FranceFranceEquityInvestissementFCPNO...2018-10-180.00-22.083-22.0830.00-25227.40-25227.400.00-25227.40-25227.40
4003166166406533FranceFranceEquityInvestissementFCPNO...2019-04-080.00-465.992-465.9920.00-563775.76-563775.760.00-563775.76-563775.76
\n", + "

5 rows × 24 columns

\n", + "
" + ], + "text/plain": [ + " Agreement - Code Company - Id Company - Ultimate Parent Id \\\n", + "0 003 166 166 \n", + "1 003 166 166 \n", + "2 003 166 166 \n", + "3 003 166 166 \n", + "4 003 166 166 \n", + "\n", + " Registrar Account - ID Registrar Account - Region \\\n", + "0 200127202 France \n", + "1 406533 France \n", + "2 406533 France \n", + "3 406533 France \n", + "4 406533 France \n", + "\n", + " RegistrarAccount - Country Product - Asset Type Product - Strategy \\\n", + "0 France Equity Investissement \n", + "1 France Diversified Patrimoine \n", + "2 France Equity Investissement \n", + "3 France Equity Investissement \n", + "4 France Equity Investissement \n", + "\n", + " Product - Legal Status Product - Is Dedie ? ... Centralisation Date \\\n", + "0 SICAV NO ... 2020-11-05 \n", + "1 FCP NO ... 2015-03-09 \n", + "2 FCP NO ... 2016-10-26 \n", + "3 FCP NO ... 2018-10-18 \n", + "4 FCP NO ... 2019-04-08 \n", + "\n", + " Quantity - Subscription Quantity - Redemption Quantity - NetFlows \\\n", + "0 1636.00 0.000 1636.000 \n", + "1 144.69 0.000 144.690 \n", + "2 0.00 -8.321 -8.321 \n", + "3 0.00 -22.083 -22.083 \n", + "4 0.00 -465.992 -465.992 \n", + "\n", + " Value Ccy - Subscription Value Ccy - Redemption Value Ccy - NetFlows \\\n", + "0 280983.00 0.00 280983.00 \n", + "1 99985.13 0.00 99985.13 \n", + "2 0.00 -9384.76 -9384.76 \n", + "3 0.00 -25227.40 -25227.40 \n", + "4 0.00 -563775.76 -563775.76 \n", + "\n", + " Value € - Subscription Value € - Redemption Value € - NetFlows \n", + "0 280983.00 0.00 280983.00 \n", + "1 99985.13 0.00 99985.13 \n", + "2 0.00 -9384.76 -9384.76 \n", + "3 0.00 -25227.40 -25227.40 \n", + "4 0.00 -563775.76 -563775.76 \n", + "\n", + "[5 rows x 24 columns]" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dg.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "d4c6dc0e-a5bc-495c-a54b-96dd238f46e3", + "metadata": {}, + "outputs": [], + "source": [ + "# Filtrer les comptes techniques\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "df['Centralisation Date'] = pd.to_datetime(df['Centralisation Date'])\n", + "dg['Centralisation Date'] = pd.to_datetime(dg['Centralisation Date'])\n", + "df = df[~df['Registrar Account - ID'].isin(['Off Distribution','Private Clients'])]\n", + "dg = dg[~dg['Registrar Account - ID'].isin(['Off Distribution','Private Clients'])]" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "a19cbae9-636f-4dc7-882c-6857daca8a11", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(4880297, 18)" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "81845380-2c3f-4bf5-81d4-4a89a3732df3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(4836569, 18)" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "c7abd838-cb2d-41dc-a280-6622b0937672", + "metadata": {}, + "outputs": [], + "source": [ + "# Date de référence\n", + "\n", + "ref_date = pd.Timestamp('2025-10-31')\n", + "\n", + "df_ref = df[df['Centralisation Date'] == ref_date]" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "688f3641-67ff-49a4-b6b0-65b64b110ff3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(24736, 18)" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_ref.shape" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1efb0610-4eaf-427c-a15c-d926d423db76", + "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": "3.13.12" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Challenge_cleaning.ipynb b/Challenge_cleaning.ipynb new file mode 100644 index 0000000..897a3f2 --- /dev/null +++ b/Challenge_cleaning.ipynb @@ -0,0 +1,3110 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 2, + "id": "4287b380-5359-49c7-ab95-ad346a3ad17a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Fichiers Flows : ['projet-bdc-data/carmignac/Flows ENSAE V1 -20251027.csv', 'projet-bdc-data/carmignac/Flows ENSAE V2 -20251105.csv']\n", + "Fichiers AUM : ['projet-bdc-data/carmignac/AUM ENSAE V1 -20251027.csv', 'projet-bdc-data/carmignac/AUM ENSAE V2 -20251105.csv']\n" + ] + } + ], + "source": [ + "# Import des données\n", + "\n", + "import os\n", + "import s3fs\n", + "import pandas as pd\n", + "\n", + "s3_ENDPOINT_URL = \"https://\" + os.environ[\"AWS_S3_ENDPOINT\"]\n", + "\n", + "fs = s3fs.S3FileSystem(client_kwargs={'endpoint_url': s3_ENDPOINT_URL})\n", + "\n", + "BUCKET = \"projet-bdc-data\"\n", + "carmignac_path = \"projet-bdc-data/carmignac\"\n", + "\n", + "# Liste des fichiers FLOWS\n", + "all_files = fs.ls(carmignac_path)\n", + "flows_files = [f for f in all_files if \"Flows\" in f and f.endswith(\".csv\")]\n", + "print(\"Fichiers Flows :\", flows_files)\n", + "\n", + "# Lire tous les fichiers dans un dictionnaire\n", + "flows_data = {}\n", + "for file_path in flows_files:\n", + " with fs.open(file_path, 'r') as f:\n", + " df = pd.read_csv(f, sep=';',low_memory=False)\n", + " flows_data[os.path.basename(file_path)] = df\n", + "\n", + "\n", + "# Liste des fichiers AUM\n", + "all_files = fs.ls(carmignac_path)\n", + "aum_files = [f for f in all_files if \"AUM\" in f and f.endswith(\".csv\")]\n", + "print(\"Fichiers AUM :\", aum_files)\n", + "\n", + "# Lire tous les fichiers dans un dictionnaire\n", + "aum_data = {}\n", + "for file_path in aum_files:\n", + " with fs.open(file_path, 'r') as f:\n", + " df = pd.read_csv(f, sep=';',low_memory=False)\n", + " aum_data[os.path.basename(file_path)] = df\n", + "\n", + "df = aum_data['AUM ENSAE V2 -20251105.csv']\n", + "dg = flows_data['Flows ENSAE V2 -20251105.csv']" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "0b66aee0-b726-4a57-9461-6a4550a625a8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Agreement - CodeCompany - IdCompany - Ultimate Parent IdRegistrar Account - IDRegistrar Account - RegionRegistrarAccount - CountryProduct - Asset TypeProduct - StrategyProduct - Legal StatusProduct - Is Dedie ?Product - FundProduct - Shareclass TypeProduct - Shareclass CurrencyProduct - IsinCentralisation DateQuantity - AUMValue - AUM CCYValue - AUM €
0003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-03-3135.36824648.666624648.6666
1003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-11-3035.36822413.055322413.0553
2003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032015-12-3135.36822051.240622051.2406
3003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032016-03-3135.36821626.117321626.1173
4003166166200000647FranceFranceDiversifiedPatrimoineFCPNOCarmignac PatrimoineAEURFR00101351032016-11-3035.36822489.450222489.4502
\n", + "
" + ], + "text/plain": [ + " Agreement - Code Company - Id Company - Ultimate Parent Id \\\n", + "0 003 166 166 \n", + "1 003 166 166 \n", + "2 003 166 166 \n", + "3 003 166 166 \n", + "4 003 166 166 \n", + "\n", + " Registrar Account - ID Registrar Account - Region \\\n", + "0 200000647 France \n", + "1 200000647 France \n", + "2 200000647 France \n", + "3 200000647 France \n", + "4 200000647 France \n", + "\n", + " RegistrarAccount - Country Product - Asset Type Product - Strategy \\\n", + "0 France Diversified Patrimoine \n", + "1 France Diversified Patrimoine \n", + "2 France Diversified Patrimoine \n", + "3 France Diversified Patrimoine \n", + "4 France Diversified Patrimoine \n", + "\n", + " Product - Legal Status Product - Is Dedie ? Product - Fund \\\n", + "0 FCP NO Carmignac Patrimoine \n", + "1 FCP NO Carmignac Patrimoine \n", + "2 FCP NO Carmignac Patrimoine \n", + "3 FCP NO Carmignac Patrimoine \n", + "4 FCP 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", + "\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 " + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "a5509e6e-ff10-4388-9fee-5cd49d01b60a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Agreement - CodeCompany - IdCompany - Ultimate Parent IdRegistrar Account - IDRegistrar Account - RegionRegistrarAccount - CountryProduct - Asset TypeProduct - StrategyProduct - Legal StatusProduct - Is Dedie ?...Centralisation DateQuantity - SubscriptionQuantity - RedemptionQuantity - NetFlowsValue Ccy - SubscriptionValue Ccy - RedemptionValue Ccy - NetFlowsValue € - SubscriptionValue € - RedemptionValue € - NetFlows
0003166166200127202FranceFranceEquityInvestissementSICAVNO...2020-11-051636.000.0001636.000280983.000.00280983.00280983.000.00280983.00
1003166166406533FranceFranceDiversifiedPatrimoineFCPNO...2015-03-09144.690.000144.69099985.130.0099985.1399985.130.0099985.13
2003166166406533FranceFranceEquityInvestissementFCPNO...2016-10-260.00-8.321-8.3210.00-9384.76-9384.760.00-9384.76-9384.76
3003166166406533FranceFranceEquityInvestissementFCPNO...2018-10-180.00-22.083-22.0830.00-25227.40-25227.400.00-25227.40-25227.40
4003166166406533FranceFranceEquityInvestissementFCPNO...2019-04-080.00-465.992-465.9920.00-563775.76-563775.760.00-563775.76-563775.76
\n", + "

5 rows × 24 columns

\n", + "
" + ], + "text/plain": [ + " Agreement - Code Company - Id Company - Ultimate Parent Id \\\n", + "0 003 166 166 \n", + "1 003 166 166 \n", + "2 003 166 166 \n", + "3 003 166 166 \n", + "4 003 166 166 \n", + "\n", + " Registrar Account - ID Registrar Account - Region \\\n", + "0 200127202 France \n", + "1 406533 France \n", + "2 406533 France \n", + "3 406533 France \n", + "4 406533 France \n", + "\n", + " RegistrarAccount - Country Product - Asset Type Product - Strategy \\\n", + "0 France Equity Investissement \n", + "1 France Diversified Patrimoine \n", + "2 France Equity Investissement \n", + "3 France Equity Investissement \n", + "4 France Equity Investissement \n", + "\n", + " Product - Legal Status Product - Is Dedie ? ... Centralisation Date \\\n", + "0 SICAV NO ... 2020-11-05 \n", + "1 FCP NO ... 2015-03-09 \n", + "2 FCP NO ... 2016-10-26 \n", + "3 FCP NO ... 2018-10-18 \n", + "4 FCP NO ... 2019-04-08 \n", + "\n", + " Quantity - Subscription Quantity - Redemption Quantity - NetFlows \\\n", + "0 1636.00 0.000 1636.000 \n", + "1 144.69 0.000 144.690 \n", + "2 0.00 -8.321 -8.321 \n", + "3 0.00 -22.083 -22.083 \n", + "4 0.00 -465.992 -465.992 \n", + "\n", + " Value Ccy - Subscription Value Ccy - Redemption Value Ccy - NetFlows \\\n", + "0 280983.00 0.00 280983.00 \n", + "1 99985.13 0.00 99985.13 \n", + "2 0.00 -9384.76 -9384.76 \n", + "3 0.00 -25227.40 -25227.40 \n", + "4 0.00 -563775.76 -563775.76 \n", + "\n", + " Value € - Subscription Value € - Redemption Value € - NetFlows \n", + "0 280983.00 0.00 280983.00 \n", + "1 99985.13 0.00 99985.13 \n", + "2 0.00 -9384.76 -9384.76 \n", + "3 0.00 -25227.40 -25227.40 \n", + "4 0.00 -563775.76 -563775.76 \n", + "\n", + "[5 rows x 24 columns]" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dg.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "d4c6dc0e-a5bc-495c-a54b-96dd238f46e3", + "metadata": {}, + "outputs": [], + "source": [ + "# Filtrer les comptes techniques\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "df['Centralisation Date'] = pd.to_datetime(df['Centralisation Date'])\n", + "dg['Centralisation Date'] = pd.to_datetime(dg['Centralisation Date'])\n", + "df = df[~df['Registrar Account - ID'].isin(['Off Distribution','Private Clients', 'Private Client'])]\n", + "dg = dg[~dg['Registrar Account - ID'].isin(['Off Distribution','Private Clients','Private Client'])]" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "c7abd838-cb2d-41dc-a280-6622b0937672", + "metadata": {}, + "outputs": [], + "source": [ + "# Date de référence\n", + "\n", + "ref_date = pd.Timestamp('2025-10-31')\n", + "\n", + "df_ref = df[df['Centralisation Date'] == ref_date]" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "1efb0610-4eaf-427c-a15c-d926d423db76", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Registrar Account - IDValue - AUM €
38904203501.623308e+09
26223647651.383209e+09
9562001274548.784361e+08
25983129338.379604e+08
10992001278098.342839e+08
\n", + "
" + ], + "text/plain": [ + " Registrar Account - ID Value - AUM €\n", + "3890 420350 1.623308e+09\n", + "2622 364765 1.383209e+09\n", + "956 200127454 8.784361e+08\n", + "2598 312933 8.379604e+08\n", + "1099 200127809 8.342839e+08" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "aum_account = (\n", + " df_ref\n", + " .groupby('Registrar Account - ID')['Value - AUM €']\n", + " .sum()\n", + " .reset_index()\n", + " .sort_values(by='Value - AUM €', ascending=False)\n", + ")\n", + "aum_account.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "1e359eff-0fa0-410d-b7cc-03b37898890b", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(431, 2)" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Garder les comptes > 5M€\n", + "aum_account = aum_account[aum_account['Value - AUM €'] > 5_000_000]\n", + "aum_account.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "99dcd4c5-56ad-4edb-90a4-ab6482455526", + "metadata": {}, + "outputs": [], + "source": [ + "# Calcul des poids\n", + "total_aum = aum_account['Value - AUM €'].sum()\n", + "\n", + "aum_account['weight'] = aum_account['Value - AUM €'] / total_aum" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "829a8019-1761-44ec-b2a4-5c4084ae0115", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Registrar Account - IDValue - AUM €weight
38904203501.623308e+090.048394
26223647651.383209e+090.041236
9562001274548.784361e+080.026188
25983129338.379604e+080.024981
10992001278098.342839e+080.024872
\n", + "
" + ], + "text/plain": [ + " Registrar Account - ID Value - AUM € weight\n", + "3890 420350 1.623308e+09 0.048394\n", + "2622 364765 1.383209e+09 0.041236\n", + "956 200127454 8.784361e+08 0.026188\n", + "2598 312933 8.379604e+08 0.024981\n", + "1099 200127809 8.342839e+08 0.024872" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "aum_account.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "12999c3a-3a6b-4035-a7c5-f230083a9143", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Score initial par compte (t = 31/10/2025) :\n" + ] + } + ], + "source": [ + "# Dictionnaire des scores initiaux\n", + "scores = aum_account.set_index('Registrar Account - ID')['weight'].to_dict()\n", + "\n", + "print(\"Score initial par compte (t = 31/10/2025) :\")\n", + "#print(scores)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "c633aa38-8d11-4c1c-897f-e29c1ceaf3dc", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "130" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dates = sorted(df['Centralisation Date'].unique(), reverse=True)\n", + "len(dates)" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "a23cd218-acf7-4143-996c-5cb209e286a9", + "metadata": {}, + "outputs": [], + "source": [ + "# Stocks : somme des Quantity par date, compte, ISIN\n", + "stocks_agg = df.groupby(['Centralisation Date', 'Registrar Account - ID', 'Product - Isin'])['Quantity - AUM'].sum()\n", + "\n", + "# Flux : somme des NetFlows par date, compte, ISIN\n", + "flows_agg = dg.groupby(['Centralisation Date', 'Registrar Account - ID', 'Product - Isin'])['Quantity - NetFlows'].sum()" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "e302da86-d226-409a-acd6-73f7965f3a60", + "metadata": {}, + "outputs": [], + "source": [ + "def check_consistency_vector(account, date_t, date_t1, tol = 0.05):\n", + " \"\"\"\n", + " Vérifie si tous les ISIN d'un compte sont cohérents :\n", + " Quantity(t) ≈ Quantity(t-1) + NetFlows\n", + " \"\"\"\n", + "\n", + " # Tous les ISIN pour ce compte sur les deux dates et flux\n", + " isins_t = stocks_agg.loc[date_t, account].index if (date_t, account) in stocks_agg.index else []\n", + " isins_t1 = stocks_agg.loc[date_t1, account].index if (date_t1, account) in stocks_agg.index else []\n", + " isins_fl = flows_agg.loc[date_t, account].index if (date_t, account) in flows_agg.index else []\n", + "\n", + " all_isins = set(isins_t) | set(isins_t1) | set(isins_fl)\n", + "\n", + " # Séries complètes avec fill_value=0 pour tous les ISIN\n", + " st_t = pd.Series({isin: stocks_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " st_t1 = pd.Series({isin: stocks_agg.get((date_t1, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " fl_t = pd.Series({isin: flows_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + "\n", + " # Vérification relative à 5 %\n", + " diff = np.abs(st_t.values - (st_t1.values + fl_t.values))\n", + " relative_error = diff / np.where(st_t.values != 0, st_t.values, 1) # éviter div par 0\n", + "\n", + " return np.all(relative_error <= tol)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "3d1357c8-50fb-4fd9-8b80-d166dcc924cb", + "metadata": {}, + "outputs": [], + "source": [ + "penalty = 0.9 # facteur de dégradation\n", + "score_history = []\n", + "\n", + "scores = aum_account.set_index('Registrar Account - ID')['weight'].to_dict()\n", + "score_history.append({\n", + " \"date\": ref_date,\n", + " \"total_score\": sum(scores.values()) # ici, scores = poids initiaux\n", + "})\n", + "\n", + "# Trier les dates décroissantes (du plus récent au plus ancien)\n", + "dates_sorted = sorted(df['Centralisation Date'].unique(), reverse=True)\n", + "\n", + "for i in range(len(dates_sorted)-1):\n", + " date_t = dates_sorted[i]\n", + " date_t1 = dates_sorted[i+1]\n", + "\n", + " new_scores = scores.copy()\n", + "\n", + " for account in scores.keys():\n", + " if not check_consistency_vector(account, date_t, date_t1):\n", + " new_scores[account] = scores[account] * penalty # dégrader le score\n", + "\n", + " scores = new_scores\n", + " score_history.append({\n", + " \"date\": date_t1,\n", + " \"total_score\": sum(scores.values())\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "f750c441-a465-42c0-ac29-1f63868eea07", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "\n", + "score_df = pd.DataFrame(score_history)\n", + "score_df = score_df.sort_values('date') # optionnel pour affichage\n", + "\n", + "plt.plot(score_df['date'], score_df['total_score'], marker='o')\n", + "plt.title(\"Évolution de la Somme des Scores avec pénalité constante\")\n", + "plt.xlabel(\"Date\")\n", + "plt.ylabel(\"Total Score\")\n", + "plt.gca().invert_xaxis() # remonter dans le temps à gauche\n", + "plt.grid(True)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "844508d8-456c-473a-86c8-aad51559794c", + "metadata": {}, + "outputs": [], + "source": [ + "# Autre score\n", + "\n", + "def score_decay_vector(account, date_t, date_t1, tol=0.05, alpha=1.0):\n", + " \"\"\"\n", + " Score diminue proportionnellement à l'erreur relative au-delà de tolérance,\n", + " en évitant les warnings si pas de données.\n", + " \"\"\"\n", + " # Tous les ISIN pour ce compte\n", + " isins_t = stocks_agg.loc[date_t, account].index if (date_t, account) in stocks_agg.index else []\n", + " isins_t1 = stocks_agg.loc[date_t1, account].index if (date_t1, account) in stocks_agg.index else []\n", + " isins_fl = flows_agg.loc[date_t, account].index if (date_t, account) in flows_agg.index else []\n", + "\n", + " all_isins = set(isins_t) | set(isins_t1) | set(isins_fl)\n", + " if len(all_isins) == 0:\n", + " return 1.0 # pas de données → pas de dégradation\n", + "\n", + " # Séries complètes float + fillna\n", + " st_t = pd.Series({isin: stocks_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " st_t1 = pd.Series({isin: stocks_agg.get((date_t1, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " fl_t = pd.Series({isin: flows_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + "\n", + " # Calcul erreur relative\n", + " denom = np.where(st_t.values != 0, st_t.values, 1)\n", + " diff = np.abs(st_t.values - (st_t1.values + fl_t.values))\n", + " relative_error = diff / denom\n", + "\n", + " # Décroissance proportionnelle\n", + " relative_error_clip = np.maximum(relative_error - tol, 0)\n", + " mean_error = np.nanmean(relative_error_clip) # remplacer NaN par 0\n", + " decay_factor = max(0, 1 - alpha * mean_error)\n", + "\n", + " return decay_factor" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "8ec0295d-add3-4674-a657-13bba49f09bf", + "metadata": {}, + "outputs": [], + "source": [ + "# Score initial = poids des comptes à la date de référence\n", + "score_history = []\n", + "scores = aum_account.set_index('Registrar Account - ID')['weight'].to_dict()\n", + "score_history.append({\n", + " \"date\": ref_date,\n", + " \"total_score\": sum(scores.values())\n", + "})\n", + "\n", + "dates_sorted = sorted(df['Centralisation Date'].unique(), reverse=True)\n", + "\n", + "alpha = 0.01 # sensibilité, 1 = décroissance égale à l'écart moyen au-delà de tolérance\n", + "tol = 0.05 # tolérance 5%\n", + "\n", + "for i in range(1, len(dates_sorted)):\n", + " date_t = dates_sorted[i-1]\n", + " date_t1 = dates_sorted[i]\n", + "\n", + " new_scores = scores.copy()\n", + "\n", + " for account in scores.keys():\n", + " decay = score_decay_vector_max(account, date_t, date_t1, tol=tol, alpha=alpha)\n", + " new_scores[account] = scores[account] * decay\n", + "\n", + " scores = new_scores\n", + " score_history.append({\n", + " \"date\": date_t1,\n", + " \"total_score\": sum(scores.values())\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": 136, + "id": "edadf900-04e6-4fca-8e58-a5ed38cfd2e6", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlQAAAHHCAYAAAB5gsZZAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAbbFJREFUeJzt3XlcVFX/B/DPsO9u7EiyKJiioJimpaaiIGpqmttjKirlQopUPlKkYpbVo4S5pPVLM6lc0mgFRcotTU3FMlxwS1NANJFN2eb8/qCZZpgBZmBgBvi8Xy9fNXfOvffcw3Dnyznfe45ECCFARERERLVmpO8KEBERETV2DKiIiIiI6ogBFREREVEdMaAiIiIiqiMGVERERER1xICKiIiIqI4YUBERERHVEQMqIiIiojpiQEVERHX222+/YenSpbhx44a+q6K1Tz/9FGvXrtV3NaiRY0ClIxKJBEuXLtXpMT/55BNIJBJcu3ZNp8etC11fp4eHB6ZNm6az45FmDPGzRY3X/fv3MXr0aNy7dw/u7u71fr7K96G6fJ6/+eYbzJo1C927d9ddBZuYadOmwcPDQ9/V0Jq6etfHd7VMkwqoZL9UVf375Zdf9F1Ftd566y0kJibquxpNQklJCVavXo1u3brBzs4OLVu2ROfOnfH888/j/Pnz+q5es3f48GEMHToUbm5usLCwwCOPPIIRI0bg888/13fVqA7CwsLQrVs3vPfee/quitz69evxySefVFvm2rVrmDFjBj777DP06dOnYSpmoG7duoWlS5ciLS1N31VptEz0XYH6sGzZMnh6eqpsb9++vR5qU7O33noLY8eOxahRo5S2P/fcc5gwYQLMzc31U7FGaMyYMUhKSsLEiRMRHh6O0tJSnD9/Ht999x369OmDjh076ruKzdbOnTsxfvx4BAQEYP78+WjVqhWuXr2KgwcP4qOPPsKkSZP0XUWqhWvXrqFHjx6IioqCkZF+/kZXd69cv3497O3tq+0BT0tLw8aNGzF69OgGqKVhu3XrFmJjY+Hh4YGAgACl9z766CNIpVL9VKwRaZIB1dChQ9GjRw99V6POjI2NYWxsrO9qNBonTpzAd999hzfffBOvvvqq0ntr165Fbm6ufipGAIClS5eiU6dO+OWXX2BmZqb03u3btxusHkIIPHz4EJaWlg12zqbMw8ND5fetodX2Xln5j9j6VlhYCGtr6wY9Z03KyspqDJZMTU0bqDaNW5Ma8tNEaWkpWrdujbCwMJX38vLyYGFhgZdfflm+7fbt25gxYwacnJxgYWEBf39/bNmypcbzVDXmvHTpUkgkEvlriUSCwsJCbNmyRT40KfuLqqq8gPXr16Nz584wNzeHq6sr5s6dqxIsPPXUU/Dz80N6ejoGDBgAKysruLm54d13362x7gBQXFyMBQsWwMHBAba2tnj66afx119/qS178+ZNTJ8+HU5OTjA3N0fnzp2xadMmjc5T2d9//42XX34ZXbp0gY2NDezs7DB06FCcOXOmxn0vX74MAHjiiSdU3jM2NkabNm2Utp0+fRpDhw6FnZ0dbGxsMGjQIJVhYdnP4PDhw5g3bx4cHBzQsmVLvPDCCygpKUFubi6mTJmCVq1aoVWrVli4cCGEEPL9r127BolEgpUrV2LdunXw8vKClZUVhgwZghs3bkAIgTfeeANt27aFpaUlRo4cib///lul/klJSejbty+sra1ha2uLYcOG4Y8//tCoTf/44w8MHDgQlpaWaNu2LZYvX17lDVST82RlZSEsLAxt27aFubk5XFxcMHLkyBrzVy5fvozHHntMJZgCAEdHR6XXUqkUq1evRpcuXWBhYQEHBweEhITg119/lZcpKyvDG2+8AW9vb5ibm8u/2IuLi5WO5eHhgeHDh2PPnj3o0aMHLC0tsXHjRgBAbm4uIiMj4e7uDnNzc7Rv3x7vvPOOSvts27YNgYGBsLW1hZ2dHbp06YLVq1dXe70AsHLlSvTp0wdt2rSBpaUlAgMD8eWXXyqV8fPzw4ABA1T2lUqlcHNzw9ixY5W2xcfHo3PnzrCwsICTkxNeeOEF3Lt3T2X/pKQk9O/fX17nxx57rMahVdn96fz58xg3bhzs7OzQpk0bzJ8/Hw8fPlQpn5CQgMDAQFhaWqJ169aYMGGCSlK6pveikpISLF68GIGBgWjRogWsra3Rt29f/PTTT9XWGVC9V3p4eOCPP/7AgQMH5PfVp556Sl5e05+7OrLP0969exEQEAALCwt06tQJu3fvVlunAwcOYM6cOXB0dETbtm3l72tzHz958iT69OkDS0tLeHp6YsOGDSr10uS7SvF+FB8fL//dWb9+PR577DEAFcO3sjaTDZmq+z4rLCzESy+9JG9DX19frFy5Uun+B1R8x0VERCAxMRF+fn7y74jk5GSlcrLP3qVLlzBt2jS0bNkSLVq0QFhYGIqKilSuV5PPnqZ09h0mmpDNmzcLAGLfvn0iJydH6d+dO3fk5aZPny5atmwpiouLlfbfsmWLACBOnDghhBCiqKhIPProo8LU1FQsWLBAvP/++6Jv374CgIiPj1faF4BYsmSJ/PXUqVNFu3btVOq4ZMkSodjsW7duFebm5qJv375i69atYuvWreLIkSNK13P16lWV/YOCgsSaNWtERESEMDY2Fo899pgoKSmRl+vfv79wdXUV7u7uYv78+WL9+vVi4MCBAoD44YcfamzLyZMnCwBi0qRJYu3ateKZZ54RXbt2VbnOrKws0bZtW+Hu7i6WLVsmPvjgA/H0008LAOK9996r8Tzt2rUTU6dOlb8+ceKE8Pb2FosWLRIbN24Uy5YtE25ubqJFixbi5s2b1R7ryJEjAoAIDw8XpaWl1ZY9e/assLa2Fi4uLuKNN94Qb7/9tvD09BTm5ubil19+kZeT/QwCAgJESEiIWLdunXjuuecEALFw4ULx5JNPikmTJon169eL4cOHCwBiy5Yt8v2vXr0q379Tp04iLi5OxMTECDMzM/H444+LV199VfTp00e8//77Yt68eUIikYiwsDClun766adCIpGIkJAQsWbNGvHOO+8IDw8P0bJlS6XPhjqZmZnCwcFBtGrVSixdulT873//Ex06dJD/LBX31/Q8ffr0ES1atBAxMTHi//7v/8Rbb70lBgwYIA4cOFBtXXx8fIS7u7u4ceNGteWEEGLatGkCgBg6dKiIj48XK1euFCNHjhRr1qyRl5k6daoAIMaOHSvWrVsnpkyZIgCIUaNGKR2rXbt2on379qJVq1Zi0aJFYsOGDeKnn34ShYWFomvXrqJNmzbi1VdfFRs2bBBTpkwREolEzJ8/X77/3r17BQAxaNAgsW7dOrFu3ToREREhnn322Rqvo23btmLOnDli7dq1Ii4uTvTs2VMAEN999528zLJly4SRkZHIzMxU2vfAgQMCgNi5c6d828yZM4WJiYkIDw8XGzZsEP/973+FtbW1yu//5s2bhUQiEX5+fuLNN98U69atEzNnzhTPPfdctfWV3V+6dOkiRowYIdauXSu/F1Ted/ny5UIikYjx48eL9evXi9jYWGFvby88PDzEvXv35OU0vRfl5OQIFxcXERUVJT744APx7rvvCl9fX2FqaipOnz6tdO7K96HK98qvvvpKtG3bVnTs2FF+X927d68QQmj8c69Ku3bthI+Pj2jZsqVYtGiRiIuLE126dBFGRkbycyjWqVOnTqJ///5izZo14u2331ZqZ03v446OjiIiIkK8//774sknnxQAxMcffywvp+l3lex+1KlTJ+Hl5SXefvtt8d5774lr166JZcuWCQDi+eefl7fZ5cuXhRCq32dSqVQMHDhQSCQSMXPmTLF27VoxYsQIAUBERkaq/Kz8/f3l99r4+Hjh5eUlrKyslL6XZW3SrVs38cwzz4j169eLmTNnyu+1ijT97Kn7Htb1d5jSsbUqbeBkH2B1/8zNzeXl9uzZIwCIb7/9Vmn/0NBQ4eXlJX8dHx8vAIiEhAT5tpKSEtG7d29hY2Mj8vLy5NtrG1AJIYS1tbVSUFH5emQ3idu3bwszMzMxZMgQUV5eLi+3du1aAUBs2rRJvq1///4CgPj000/l24qLi4Wzs7MYM2aMyrkUpaWlCQBizpw5StsnTZqkcp0zZswQLi4uSr8YQggxYcIE0aJFC1FUVFTtuSoHVA8fPlS6NiEqbgLm5uZi2bJl1R5LKpXKr9vJyUlMnDhRrFu3Tvz5558qZUeNGiXMzMzkNwwhhLh165awtbUV/fr1k2+T/QyCg4OFVCqVb+/du7eQSCRi1qxZ8m1lZWWibdu2on///kp1ByAcHBxEbm6ufHt0dLT8RqMY/E2cOFGYmZmJhw8fCiGEyM/PFy1bthTh4eFK9c/KyhItWrRQ2V5ZZGSkACCOHTsm33b79m3RokULpc+Wpue5d++eACD+97//VXtedT7++GMBQJiZmYkBAwaI119/XRw6dEjl5/3jjz8KAGLevHkqx5D9DGSf0ZkzZyq9//LLLwsA4scff5Rva9eunQAgkpOTlcq+8cYbwtraWly8eFFp+6JFi4SxsbG4fv26EEKI+fPnCzs7O1FWVqb1NVf+/JeUlAg/Pz8xcOBA+bYLFy4IAErBohBCzJkzR9jY2MiPcejQIQFAfPbZZ0rlkpOTlbbn5uYKW1tb0atXL/HgwQOlsoqfYXVk96enn35apS4AxJkzZ4QQQly7dk0YGxuLN998U6nc77//LkxMTJS2a3ovKisrU/kj9969e8LJyUlMnz5daXtNAZUQQnTu3Fnpd1FG0597VWSfp127dsm33b9/X7i4uIhu3bqp1OnJJ59U+uzU5j6+atUq+bbi4mIREBAgHB0d5cGXpt9VsvuRnZ2duH37ttJ1nThxQgAQmzdvVrnmyt9niYmJAoBYvny5UrmxY8cKiUQiLl26JN8m+51X3HbmzBmVz7zss1f5Zz169GjRpk0b+WttPnuaBFR1/Q5T1CSH/NatW4eUlBSlf0lJSfL3Bw4cCHt7e2zfvl2+7d69e0hJScH48ePl23744Qc4Oztj4sSJ8m2mpqaYN28eCgoKcODAgYa5oH/s27cPJSUliIyMVEr+DA8Ph52dHb7//nul8jY2Npg8ebL8tZmZGXr27IkrV65Ue54ffvgBADBv3jyl7ZGRkUqvhRDYtWsXRowYASEE7ty5I/8XHByM+/fv49SpU1pdo7m5ufzaysvLcffuXdjY2MDX17fGY0kkEuzZswfLly9Hq1at8MUXX2Du3Llo164dxo8fL+9OLy8vx969ezFq1Ch4eXnJ93dxccGkSZNw+PBh5OXlKR17xowZSkO1vXr1ghACM2bMkG8zNjZGjx491Lbvs88+ixYtWijtDwCTJ0+GiYmJ0vaSkhLcvHkTAJCSkoLc3FxMnDhRqX2NjY3Rq1evGodDfvjhBzz++OPo2bOnfJuDgwP+85//KJXT9DyWlpYwMzPD/v371Q4zVWf69OlITk7GU089hcOHD+ONN95A37590aFDBxw5ckRebteuXZBIJFiyZInKMWQ/A9lnNCoqSun9l156CQBUfhc8PT0RHBystG3nzp3o27cvWrVqpXTNQUFBKC8vx8GDBwEALVu2RGFhIVJSUrS6XgBKeVr37t3D/fv30bdvX6XPso+PDwICApTuR+Xl5fjyyy8xYsQI+TF27tyJFi1aYPDgwUr1DQwMhI2NjfxnlJKSgvz8fCxatAgWFhZK9VH8DFdn7ty5Sq9ffPFFAP+2++7duyGVSjFu3Dilujg7O6NDhw4qn0tN7kXGxsby4WCpVIq///4bZWVl6NGjh9b3kepo+nOvjqurq1Iiu52dHaZMmYLTp08jKytLqWx4eLhSfpe293ETExO88MIL8tdmZmZ44YUXcPv2bZw8eRKA9t9VY8aMgYODQ43XWZUffvgBxsbGKt8RL730EoQQSt+3ABAUFARvb2/5665du8LOzk7tvXLWrFlKr/v27Yu7d+/K78nafvaqo+vvsCaZlN6zZ89qk9JNTEwwZswYfP755yguLoa5uTl2796N0tJSpYDqzz//RIcOHVSeXHn00Ufl7zck2fl8fX2VtpuZmcHLy0ulPm3btlW5gbZq1Qq//fZbjecxMjJS+gVQd96cnBzk5ubiww8/xIcffqj2WNomG8tyZ9avX4+rV6+ivLxc/l7lHCh1zM3N8dprr+G1115DZmYmDhw4gNWrV2PHjh0wNTVFQkICcnJyUFRUpHI9QMXPViqV4saNG+jcubN8+yOPPKJUThYcVZ5zp0WLFmoDDW32ByA/RkZGBoCKPwLUsbOzU7td5s8//5QHb4oqX7um5zE3N8c777yDl156CU5OTnj88ccxfPhwTJkyBc7OztXWBQCCg4MRHByMoqIinDx5Etu3b8eGDRswfPhwnD9/Ho6Ojrh8+TJcXV3RunXraq/LyMhI5cldZ2dntGzZUuV3Qd1TvxkZGfjtt9+q/GKRfXbnzJmDHTt2yKd7GDJkCMaNG4eQkJAar/e7777D8uXLkZaWppTbVfn3cvz48Xj11Vdx8+ZNuLm5Yf/+/bh9+7bS/SgjIwP3799XyTerXF9ZLqGfn1+N9atKhw4dlF57e3vDyMhInqOUkZEBIYRKOZnKScya3ou2bNmCVatW4fz58ygtLZVvV/fzqy1Nf+7Vad++vcr1+Pj4AKjIU1L8Xahcd23v466uriqJ7Irnevzxx7X+rqpre/75559wdXWFra2tRuerfP8DKn7+mtwrW7VqBaDinmhnZ6f1Z686uv4Oa5IBlSYmTJiAjRs3IikpCaNGjcKOHTvQsWNH+Pv76+T4Vf0lqBgg1LeqnnoRlZIGa0uWwDl58mRMnTpVbZmuXbtqdcy33noLr7/+OqZPn4433ngDrVu3hpGRESIjI7V+bNfFxQUTJkzAmDFj0LlzZ+zYsaPGeWmqUlVbqtuurn212V/xGLJr3rp1q9qARbF3qy60OU9kZCRGjBiBxMRE7NmzB6+//jpWrFiBH3/8Ed26ddPofFZWVujbty/69u0Le3t7xMbGIikpqcrPUVU07XFR90SfVCrF4MGDsXDhQrX7yL60HB0dkZaWhj179iApKQlJSUnYvHkzpkyZUu0DKocOHcLTTz+Nfv36Yf369XBxcYGpqSk2b96skhw+fvx4REdHY+fOnYiMjMSOHTvQokULpaBNKpXC0dERn332mdrz1aXHoSaV21kqlUIikSApKUntZ9jGxkbptSb3ooSEBEybNg2jRo3CK6+8AkdHRxgbG2PFihXyIFEXNP2564ohPk3a0HXS5rtIk3uiNp+96uj6O6zZBlT9+vWDi4sLtm/fjieffBI//vgjXnvtNaUy7dq1w2+//QapVKoU+csmiGzXrl2Vx2/VqpXax/TV9Wpp+qUgO9+FCxeUhqpKSkpw9epVBAUFaXQcTc4jlUpx+fJlpb+iLly4oFRO9gRgeXm5zs795ZdfYsCAAfj444+Vtufm5sLe3r5WxzQ1NUXXrl2RkZGBO3fuwMHBAVZWVirXA1T8bI2MjBpktmdNyHoJHR0da9XG7dq1k/c+Kap87dqex9vbGy+99BJeeuklZGRkICAgAKtWrUJCQoLWdZT1JmdmZsqPvWfPHvz9999V9lLJPqMZGRnyv4oBIDs7G7m5udX+bipeQ0FBgUbXa2ZmhhEjRmDEiBGQSqWYM2cONm7ciNdff73K+e127doFCwsL7NmzR2l+pM2bN6uU9fT0RM+ePbF9+3ZERERg9+7dGDVqlNJ+3t7e2LdvH5544olqvxBlP8uzZ8/Weu69jIwMpV6MS5cuQSqVyp/08vb2hhACnp6eOgtAvvzyS3h5eWH37t1K90R1Q7+aqOq+qs3PvSqXLl2CEELpHBcvXgSAGmcU1/Y+fuvWLZXpFiqfqy7fVTKafg/Jjrdv3z7k5+cr9VJpc77a0uVnT9ffYU0yh0oTRkZGGDt2LL799lts3boVZWVlSt3rABAaGoqsrCyl3IaysjKsWbMGNjY26N+/f5XH9/b2xv3795W6tDMzM/HVV1+plLW2ttZojqSgoCCYmZnh/fffV4rsP/74Y9y/fx/Dhg2r8RiaGDp0KADg/fffV9oeHx+v9NrY2BhjxozBrl27cPbsWZXj5OTkaH1uY2Njlb9adu7cKc8pqk5GRgauX7+usj03NxdHjx5Fq1at4ODgAGNjYwwZMgRff/210qP+2dnZ+Pzzz/Hkk0/WOJTWUIKDg2FnZ4e33npLaQhEpqY2Dg0NxS+//ILjx48r7VO5l0PT8xQVFak8Pu/t7Q1bW1uV6QoqS01NVbtdlpcjC97HjBkDIQRiY2NVyso+G6GhoQBUP5NxcXEAoNHvwrhx43D06FHs2bNH5b3c3FyUlZUBAO7evav0npGRkfyv1uqu2djYGBKJRKlX+tq1a1WuijB+/Hj88ssv2LRpE+7cuaNyPxo3bhzKy8vxxhtvqOxbVlYmv4cMGTIEtra2WLFihcrPStPe6XXr1im9XrNmDYB/7w3PPPMMjI2NERsbq3JMIYRKm2lC1tugeLxjx47h6NGjWh8LqPq+qunPvTq3bt1Supfn5eXh008/RUBAQI1D39rex8vKyuTTfAAVgdfGjRvh4OCAwMBAAHX7rpKRBWyafBeFhoaivLxcZf3D9957DxKJRP45qQ+6/Ozp+jusSfZQJSUlqV1mpE+fPkp/EYwfPx5r1qzBkiVL0KVLF6W/dAHg+eefx8aNGzFt2jScPHkSHh4e+PLLL/Hzzz8jPj5eZfxY0YQJE/Df//4Xo0ePxrx581BUVIQPPvgAPj4+KklugYGB2LdvH+Li4uDq6gpPT0+1eS8ODg6Ijo5GbGwsQkJC8PTTT+PChQvyOUQUkz7rIiAgABMnTsT69etx//599OnTB6mpqbh06ZJK2bfffhs//fQTevXqhfDwcHTq1Al///03Tp06hX379qmdU6k6w4cPx7JlyxAWFoY+ffrg999/x2effab0c6vKmTNnMGnSJAwdOhR9+/ZF69atcfPmTWzZsgW3bt1CfHy8/Ka9fPlypKSk4Mknn8ScOXNgYmKCjRs3ori4WOO5uhqCnZ0dPvjgAzz33HPo3r07JkyYAAcHB1y/fh3ff/89nnjiiWoXdV24cCG2bt2KkJAQzJ8/H9bW1vjwww/lf9Fqe56LFy9i0KBBGDduHDp16gQTExN89dVXyM7OxoQJE6q9lpEjR8LT0xMjRoyAt7c3CgsLsW/fPnz77bd47LHHMGLECADAgAED8Nxzz+H9999HRkYGQkJCIJVKcejQIQwYMAARERHw9/fH1KlT8eGHHyI3Nxf9+/fH8ePHsWXLFowaNUrtvE6VvfLKK/jmm28wfPhwTJs2DYGBgSgsLMTvv/+OL7/8EteuXYO9vT1mzpyJv//+GwMHDkTbtm3x559/Ys2aNQgICFC5ZygaNmwY4uLiEBISgkmTJuH27dtYt24d2rdvrzaPcdy4cXj55Zfx8ssvo3Xr1ip/Mffv3x8vvPACVqxYgbS0NAwZMgSmpqbIyMjAzp07sXr1aowdOxZ2dnZ47733MHPmTDz22GOYNGkSWrVqhTNnzqCoqEijefSuXr2Kp59+GiEhITh69CgSEhIwadIkeUqEt7c3li9fjujoaFy7dg2jRo2Cra0trl69iq+++grPP/+80nx+mhg+fDh2796N0aNHY9iwYbh69So2bNiATp06oaCgQKtjARX31Q8++ADLly9H+/bt4ejoiIEDB2r8c6+Oj48PZsyYgRMnTsDJyQmbNm1Cdna22t7HyrS9j7u6uuKdd97BtWvX4OPjg+3btyMtLQ0ffvihPF+oLt9VMt7e3mjZsiU2bNgAW1tbWFtbo1evXmrzrUaMGIEBAwbgtddew7Vr1+Dv74+9e/fi66+/RmRkpEr+rS7p+rOn0+8wjZ8HbASqmzYBah4HlUqlwt3dXe3jnzLZ2dkiLCxM2NvbCzMzM9GlSxe1j5Wi0qOYQlTMX+Pn5yfMzMyEr6+vSEhIUDttwvnz50W/fv2EpaWlACCfRkDdo8BCVDxe27FjR2FqaiqcnJzE7NmzlebeEKLicdvOnTur1LOq6Rwqe/DggZg3b55o06aNsLa2FiNGjBA3btxQe53Z2dli7ty5wt3dXZiamgpnZ2cxaNAg8eGHH9Z4HnXTJrz00kvCxcVFWFpaiieeeEIcPXpU9O/fX+0j0JXr8fbbb4v+/fsLFxcXYWJiIlq1aiUGDhwovvzyS5Xyp06dEsHBwcLGxkZYWVmJAQMGyOcAk5H9DGRzk8nIfo45OTlK26dOnSqsra3lr2WPKVeeZuCnn35SmWOouvP99NNPIjg4WLRo0UJYWFgIb29vMW3aNPHrr79W2yZCCPHbb7+J/v37CwsLC+Hm5ibeeOMN+RQGlT9bNZ3nzp07Yu7cuaJjx47C2tpatGjRQvTq1Uvs2LGjxnp88cUXYsKECcLb21tYWloKCwsL0alTJ/Haa68pTUEiRMUj9P/73/9Ex44dhZmZmXBwcBBDhw4VJ0+elJcpLS0VsbGxwtPTU5iamgp3d3cRHR0tn3JCpl27dmLYsGFq65Sfny+io6NF+/bthZmZmbC3txd9+vQRK1eulD+S/uWXX4ohQ4YIR0dHYWZmJh555BHxwgsvqMwbpc7HH38sOnToIMzNzUXHjh3F5s2b1d4DZJ544gm100Eo+vDDD0VgYKCwtLQUtra2okuXLmLhwoXi1q1bSuW++eYb0adPH2FpaSns7OxEz549xRdffFFtfWV1S09PF2PHjhW2traiVatWIiIiQmUKBiGE2LVrl3jyySeFtbW1sLa2Fh07dhRz584VFy5ckJfR9F4klUrFW2+9Jdq1ayfMzc1Ft27dxHfffafRo+/q7pVZWVli2LBhwtbWVgBQun9o8nOviuzztGfPHtG1a1f5z1bT32UZbe7jv/76q+jdu7ewsLAQ7dq1E2vXrlU5nibfVVXdj2S+/vpr0alTJ2FiYqL0nanuZ5Cfny8WLFggXF1dhampqejQoYP43//+pzI1BwAxd+5clXNVvvdXdU+t6ntQk8+eJp8dIer2HaZI8s8JiIiomVu6dCliY2ORk5NT65zFps7DwwN+fn747rvv6v1cTz31FO7cuaN2OIoMT7PNoSIiIiLSFQZURERERHXEgIqIiIiojphDRURERFRH7KEiIiIiqiMGVERERER11CQn9qwrqVSKW7duwdbWVqvp+ImIiEh/hBDIz8+Hq6urymLR9Y0BlRq3bt0ymLXciIiISDs3btxA27ZtG/ScDKjUkE3Tf+PGjVqt6VZaWoq9e/fKl4agumF76g7bUrfYnrrF9tSt5tieeXl5cHd312i5HV1jQKWGbJjPzs6u1gGVlZUV7Ozsms2HuD6xPXWHbalbbE/dYnvqVnNuT32k6zApnYiIiKiOGFARERER1REDKiIiIqI6YkBFREREVEcMqIiIiIjqiAEVERERUR0xoCIiIiKqIwZURERERHXEgIqIiIiojhhQEREREdWRXgOqgwcPYsSIEXB1dYVEIkFiYmKN++zfvx/du3eHubk52rdvj08++USlzLp16+Dh4QELCwv06tULx48f133liYiIiP6h14CqsLAQ/v7+WLdunUblr169imHDhmHAgAFIS0tDZGQkZs6ciT179sjLbN++HVFRUViyZAlOnToFf39/BAcH4/bt2/V1GRpLPpuJkPiD8I1JQkj8QSSfzdR3lYiIiEgH9BpQDR06FMuXL8fo0aM1Kr9hwwZ4enpi1apVePTRRxEREYGxY8fivffek5eJi4tDeHg4wsLC0KlTJ2zYsAFWVlbYtGlTfV2GRpLPZmJWwilcyMpHcZkUF7LyMSvhFIMqIiKiJsBE3xXQxtGjRxEUFKS0LTg4GJGRkQCAkpISnDx5EtHR0fL3jYyMEBQUhKNHj1Z53OLiYhQXF8tf5+XlAahYqbu0tFTresr2Udz3vZSLkAAQ/7yW/Xd2win4ONngxQHeCO7spPW5mgN17Um1w7bULbanbrE9das5tqc+r7VRBVRZWVlwclIOOpycnJCXl4cHDx7g3r17KC8vV1vm/PnzVR53xYoViI2NVdm+d+9eWFlZ1bq+KSkp8v+/fNsYAhKVMgLAhex8RGw7g+k+5fBvI1TKUAXF9qS6YVvqFttTt9ieutWc2rOoqEhv525UAVV9iY6ORlRUlPx1Xl4e3N3dMWTIENjZ2Wl9vNLSUqSkpGDw4MEwNTUFAKy/cgQXswugPlySQALgsyumSLgCeLaxYo+VAnXtSbXDttQttqdusT11qzm2p2yESR8aVUDl7OyM7OxspW3Z2dmws7ODpaUljI2NYWxsrLaMs7Nzlcc1NzeHubm5ynZTU9M6fQgV918w2AezEk5BIgGEmqhKACgukwIALmYXIGLbGZgYSdDe0QaRQR0Q4udS63o0FXX9edC/2Ja6xfbULbanbjWn9tTndTaqeah69+6N1NRUpW0pKSno3bs3AMDMzAyBgYFKZaRSKVJTU+Vl9CXEzwUbJndHR2dbNQN/ymTxVplUMHmdiIioEdBrD1VBQQEuXbokf3316lWkpaWhdevWeOSRRxAdHY2bN2/i008/BQDMmjULa9euxcKFCzF9+nT8+OOP2LFjB77//nv5MaKiojB16lT06NEDPXv2RHx8PAoLCxEWFtbg11dZiJ8LQvxc5E/8VdVbpUgAkACYvy0NQBo87a3ZY0VERGRg9NpD9euvv6Jbt27o1q0bgIpgqFu3bli8eDEAIDMzE9evX5eX9/T0xPfff4+UlBT4+/tj1apV+L//+z8EBwfLy4wfPx4rV67E4sWLERAQgLS0NCQnJ6skquuTYm+VuYkRzE2q/zHIhgM53QIREZFh0msP1VNPPQVRTReNulnQn3rqKZw+fbra40ZERCAiIqKu1atXst4qANr3WEmA1akZ7KUiIiIyEI0qh6qpUuyxMjGqKcOqIui6klPYADUjIiIiTTSqp/yasso9VqtTM+RBk+zpP0UlZVKExB9kPhUREZEBYEBlgDQZDhSAPJ/KraUF7hSUMGGdiIhITzjkZ+Cqm25BFl/dzH3IhHUiIiI9YkDVCIT4uSBpfj+YafA0oCxhnYiIiBoOA6pGxNPeuuZJQQVwLjMfIfEH2VNFRETUQJhD1YhEBnXQeHoF5lcRERE1HPZQNSKVJwRt28oSQMUwX2XMryIiImo47KFqZBSfAAT+nWLhXGZ+tftxQlAiIqL6wx6qRk6WsK7RosucEJSIiKheMKBqIiKDOsh7oaojmxCUQ39ERES6w4CqidA0v0pxQlAGVURERLrBHKompKr8qvOZ+VB8KFD2//O3pQFI4xOAREREdcQeqiaspglBi8ukfAKQiIhIBxhQNQM1TQgq67GanXCK+VVERES1wCG/ZkDTCUG54DIREVHtsIeqGaicsG5ezZqAnBCUiIhIe+yhaiYUE9aTz2ZqvIQNJwQlIiKqGXuomiHFHquaJgMFuOAyERFRTdhD1UzJeqy06a1ifhUREZF67KFq5rjgMhERUd2xh4rqtOAyUDHdgq+zLXuriIio2WJARSpkAVZI/EFcyFKeZV0dTrdARETNHYf8qEqaLrgMcDiQiIiaNwZUVCVt8qsq4+zrRETUnHDIj6ql6YLLVVEcDtwwuTuHAImIqEliQEVaqc10CwKABMD8bWkA0phfRURETQ6H/KhWtB0OFACKy6TMryIioiaJPVRUa7UdDuR0C0RE1NSwh4p0JsTPBUnz++GDyd0B1Jy8rphfxd4qIiJqzBhQkc5VHg40N6n6Y6a4+DIREVFjxSE/qheKw4E1JbALAVzJKWzgGhIREekOe6io3in2WFU1CujlYN2gdSIiItIlBlTUIGrKr7p17wF8Y5I4CSgRETVKDKioQVXOr2phaQoAuP+wjFMqEBFRo8UcKmpwivlVIfEHcf9Bqfw9dVMqDPK110MtiYiINMceKtKrq3fUJ6MrTqmw54/shq0UERGRlhhQkV552ltXmagum1Jh7f7LDVklIiIirTGgIr2KDOogD5zUEQK4kFWAl34xxvC1R5hbRUREBokBFemVJlMqCABlQoKL2QVMWCciIoPEgIr0TtMlazirOhERGSoGVGQwKk+poC6uEgI4l5nP+aqIiMigMKAigyLrrbqwfCh8qxkG5HxVRERkSBhQkcGqLmFdAJAAmL8tjTOsExGR3jGgIoMlGwL0dbLBv1N+/ksAKC6TcoZ1IiLSO70HVOvWrYOHhwcsLCzQq1cvHD9+vMqypaWlWLZsGby9vWFhYQF/f38kJycrlcnPz0dkZCTatWsHS0tL9OnTBydOnKjvy6B6EuLngm/n9oGrFaoc/gOYsE5ERPql14Bq+/btiIqKwpIlS3Dq1Cn4+/sjODgYt2/fVls+JiYGGzduxJo1a5Ceno5Zs2Zh9OjROH36tLzMzJkzkZKSgq1bt+L333/HkCFDEBQUhJs3bzbUZVE9CGkrrXa+KoAJ60REpD96Daji4uIQHh6OsLAwdOrUCRs2bICVlRU2bdqktvzWrVvx6quvIjQ0FF5eXpg9ezZCQ0OxatUqAMCDBw+wa9cuvPvuu+jXrx/at2+PpUuXon379vjggw8a8tJIx/zbCKyd4C9/AtDcpOqPrmz474m3U5lfRUREDUJviyOXlJTg5MmTiI6Olm8zMjJCUFAQjh49qnaf4uJiWFhYKG2ztLTE4cOHAQBlZWUoLy+vtkxVxy0uLpa/zsvLA1AxxFhaWlrVblWS7VObfUmVrB0H+rRGcGcnAMCeP7IRse0MJJKKnilFspc3cx8C+DfAMjGSwNvBGi8O8JYfp7nhZ1O32J66xfbUrebYnvq8VokQlb+OGsatW7fg5uaGI0eOoHfv3vLtCxcuxIEDB3Ds2DGVfSZNmoQzZ84gMTER3t7eSE1NxciRI1FeXi4PiPr06QMzMzN8/vnncHJywhdffIGpU6eiffv2uHDhgtq6LF26FLGxsSrbP//8c1hZWenoiknXztyVIPkvI9wqAqrPsAL+fS6w4r/Tfcrh30YvH30iIqonRUVFmDRpEu7fvw87O7sGPXejCqhycnIQHh6Ob7/9FhKJBN7e3ggKCsKmTZvw4MEDAMDly5cxffp0HDx4EMbGxujevTt8fHxw8uRJnDt3Tm1d1PVQubu7486dO7X6gZSWliIlJQWDBw+Gqamp1vuTsprac/jaI7iYXaDmOUD1JBLA18kG387to9uKNgL8bOoW21O32J661RzbMy8vD/b29noJqPQ25Gdvbw9jY2NkZ2crbc/Ozoazs7PafRwcHJCYmIiHDx/i7t27cHV1xaJFi+Dl5SUv4+3tjQMHDqCwsBB5eXlwcXHB+PHjlcpUZm5uDnNzc5XtpqamdfoQ1nV/UlZVey4Y7INZCafUDv+pI1tw2S92HzztrREZ1AEhfi71UGPDxc+mbrE9dYvtqVvNqT31eZ16S0o3MzNDYGAgUlNT5dukUilSU1OVeqzUsbCwgJubG8rKyrBr1y6MHDlSpYy1tTVcXFxw79497NmzR20ZahoqL1nTtpUlgBqeCAQ4fxUREemM3nqoACAqKgpTp05Fjx490LNnT8THx6OwsBBhYWEAgClTpsDNzQ0rVqwAABw7dgw3b95EQEAAbt68iaVLl0IqlWLhwoXyY+7ZswdCCPj6+uLSpUt45ZVX0LFjR/kxqWkK8XNR6mVKPpuJ1akZyMguQJm06m4r2TuzE07B19m2WfZWERFR3ek1oBo/fjxycnKwePFiZGVlISAgAMnJyXByqngC6/r16zAy+rcT7eHDh4iJicGVK1dgY2OD0NBQbN26FS1btpSXuX//PqKjo/HXX3+hdevWGDNmDN58881m091JFRQDLFlwdSWnECVlUrW5VgL/Pg24YXJ3BlVERKQVvQZUABAREYGIiAi17+3fv1/pdf/+/ZGenl7t8caNG4dx48bpqnrUBCgGVyHxB3EhK7/KoEq2PiCQ1mzzq4iISHt6X3qGqCFVt+AywPUBiYiodhhQUbOimMCu0cxVXB+QiIg0wICKmp0QPxckze+HDyZ3B8D1AYmIqO4YUFGzVXm6BU3WB2RQRURE6ug9KZ1Inyo/DVjVBKFMWCciouqwh4roH4o9VuowYZ2IiKrCgIpIgSy/qqakdSasExGRIgZURGrUNL0CwIR1IiL6FwMqIjWYsE5ERNpgUjpRFbRKWP9n+I9J6kREzRN7qIg0UGPCugDOZ+bDNyaJQ4BERM0QAyoiDdWUsC57CpBDgEREzQ8DKiItabIeIJ8AJCJqXhhQEWmpcsK62t4qAVzJKWzwuhERkX4woCKqBdnw34XlQ+FbxRBgSZmU+VRERM0EAyqiOqpqCFCAUyoQETUXDKiI6khxCLByT5XiGoB8ApCIqOliQEWkA7IhQDM1E4ByDUAioqaPARWRDnnaW3MNQCKiZogBFZEOcQ1AIqLmiUvPEOmQLJ9qdWqGfNqE4jKp2rKy4T+3lha4U1ACT3trRAZ14PI1RESNEHuoiHRMcUqF1RMCAKjvsZItCXgz9yHzq4iIGjkGVET1qKY1ABXJAqzZCac4HEhE1MgwoCKqZzWtAVgZ568iImp8GFARNRBNEtZl+DQgEVHjwoCKqIFUXgOwbStLANUssiyA85n5nBCUiKgR4FN+RA0oxM9F6Sm+5LOZWJ2agfOZ+fIcKkWySUFlQ4AbJnfnU4BERAaIPVREeiTLr/pgcncA1fRW/fNfJqwTERkmBlREBqDycGBVaVZMWCciMkwc8iMyEIrDgSHxB3Ehq+phQNmCy0AaJwQlIjIA7KEiMkA1PRHIBZeJiAwLAyoiA6Q4BFjTLAvMryIi0j8GVEQGStOEdRnmVxER6Q8DKiIDVzlh3dyk6l9bxfwqzl9FRNRwGFARNQKaLrgMML+KiEgfGFARNTJ1ya/a80d2fVePiKhZYkBF1AjVNr8qYtsZnLmryRLNRESkDQZURI2Y1vlVEiD5L/7aExHpGu+sRI2cVvlVArhVBAxfe4R5VUREOsSAiqgJ0Sy/SoKL2QVMVici0iEuPUPUxMiWsEk+m4lZCacgkVT0TCmSvZy3LQ0SLl9DRFRn7KEiaqIUe6uqUsLpFYiIdIIBFVETJsuvqmmKBVnC+urUjIaqGhFRk8KAiqgZqGmxZaBiWPBcZj5nVyciqgUGVETNgGz4z9fJBiYSUe30Chz+IyLSnt4DqnXr1sHDwwMWFhbo1asXjh8/XmXZ0tJSLFu2DN7e3rCwsIC/vz+Sk5OVypSXl+P111+Hp6cnLC0t4e3tjTfeeAOiclYuUTMT4ueCb+f2warHy7FqbBcA6nusOPxHRKQ9vQZU27dvR1RUFJYsWYJTp07B398fwcHBuH37ttryMTEx2LhxI9asWYP09HTMmjULo0ePxunTp+Vl3nnnHXzwwQdYu3Ytzp07h3feeQfvvvsu1qxZ01CXRWTwgjs7VZuwLgRwJaewgWtFRNR46XXahLi4OISHhyMsLAwAsGHDBnz//ffYtGkTFi1apFJ+69ateO211xAaGgoAmD17Nvbt24dVq1YhISEBAHDkyBGMHDkSw4YNAwB4eHjgiy++qLbnq7i4GMXFxfLXeXl5ACp6xEpLS7W+Ltk+tdmXVLE9dUexLQf52mOQrz2Grz2Ci9kFqNyHWyYV8IlJgmcbK7w4wBvBnZ0avsIGjp9N3WJ76lZzbE99XqveAqqSkhKcPHkS0dHR8m1GRkYICgrC0aNH1e5TXFwMCwsLpW2WlpY4fPiw/HWfPn3w4Ycf4uLFi/Dx8cGZM2dw+PBhxMXFVVmXFStWIDY2VmX73r17YWVlpe2lyaWkpNR6X1LF9tQdxbZ8ooUEF7KNIYGAgAT/DPqhXCpQLhW4kF2xBuB0n3L4t+HQuTr8bOoW21O3mlN7FhUV6e3ceguo7ty5g/Lycjg5Kf/V6+TkhPPnz6vdJzg4GHFxcejXrx+8vb2RmpqK3bt3o7y8XF5m0aJFyMvLQ8eOHWFsbIzy8nK8+eab+M9//lNlXaKjoxEVFSV/nZeXB3d3dwwZMgR2dnZaX1tpaSlSUlIwePBgmJqaar0/KWN76o66tgwF0P2PbKzdfxlX7hRBAqC4TKqwlwQSCXAkrwWin+ujj2obLH42dYvtqVvNsT1lI0z60KhmSl+9ejXCw8PRsWNHSCQSeHt7IywsDJs2bZKX2bFjBz777DN8/vnn6Ny5M9LS0hAZGQlXV1dMnTpV7XHNzc1hbm6ust3U1LROH8K67k/K2J66U7kthwe0xfCAtgAA35gklfJCAFfvFLH9q8DPpm6xPXWrObWnPq9Tb0np9vb2MDY2RnZ2ttL27OxsODs7q93HwcEBiYmJKCwsxJ9//onz58/DxsYGXl5e8jKvvPIKFi1ahAkTJqBLly547rnnsGDBAqxYsaJer4eoqfC0t1aZBFQiAbwcrPVSHyKixkBvAZWZmRkCAwORmpoq3yaVSpGamorevXtXu6+FhQXc3NxQVlaGXbt2YeTIkfL3ioqKYGSkfFnGxsaQSqWVD0NEaqibBFQIYP4gH73ViYjI0Ol12oSoqCh89NFH2LJlC86dO4fZs2ejsLBQ/tTflClTlJLWjx07ht27d+PKlSs4dOgQQkJCIJVKsXDhQnmZESNG4M0338T333+Pa9eu4auvvkJcXBxGjx7d4NdH1BgprgFobFQRVT3qYosQP/U9x0REpOccqvHjxyMnJweLFy9GVlYWAgICkJycLE9Uv379ulJv08OHDxETE4MrV67AxsYGoaGh2Lp1K1q2bCkvs2bNGrz++uuYM2cObt++DVdXV7zwwgtYvHhxQ18eUaMV4ueCED8X3Pi7CH3f/QnnMvNx4+8iuLeu/VOvRERNmd6T0iMiIhAREaH2vf379yu97t+/P9LT06s9nq2tLeLj4xEfH6+jGhI1X+6trfBke3scvnQHo9b/jIKHZfC0t0ZkUAeE+Lnou3pERAZD70vPEJFh8/1nNvW7BSUoLpNyrT8iIjUYUBFRtQ5n3FF6zbX+iIhUMaAiompdu6u6ph/X+iMiUsaAioiqpW5eKgAoKZMiJP4gh/6IiMCAiohqIJ+XqtJ2ATCfiojoHwyoiKha8nmpXGzVBlUSAPO3pcE3Jok9VkTUbDGgIqIahfi5IGl+P5iZqN4yBCoWU+YTgETUnDGgIiKNVZVPJcMnAImouWJARUQaU7fOX2V8ApCImiMGVESkMcV1/sxNjGCuZghQIgG8HKz1UDsiIv1hQEVEWpHlU11YPhSrJwQAUH4CUAggI7uACepE1KwwoCKiWlN8AtBIIaoqkwomqBNRs8KAiojqRNZj1cHRVmk7E9SJqDmpdUBVUlKCCxcuoKysTJf1IaJGikvUEFFzpnVAVVRUhBkzZsDKygqdO3fG9evXAQAvvvgi3n77bZ1XkIgaB3VTKkjABHUiah60Dqiio6Nx5swZ7N+/HxYWFvLtQUFB2L59u04rR0SNh7opFQSAvwtLOIs6ETV5WgdUiYmJWLt2LZ588klIFO6cnTt3xuXLl3VaOSJqPBSnVDAzMYKpccX9ITuvmLOoE1GTp3VAlZOTA0dHR5XthYWFSgEWETU/sgT1i8uHol0b5aE+JqkTUVOmdUDVo0cPfP/99/LXsiDq//7v/9C7d2/d1YyIGrUbfxepbGOSOhE1VSba7vDWW29h6NChSE9PR1lZGVavXo309HQcOXIEBw4cqI86ElEj5GlvjQtZ+RAK25ikTkRNldY9VE8++STOnDmDsrIydOnSBXv37oWjoyOOHj2KwMDA+qgjETVCVSWpcxZ1ImqKtAqoSktLMX36dEgkEnz00Uc4fvw40tPTkZCQgC5dutRXHYmoEVJMUjdWmEads6gTUVOkVUBlamqKXbt21VddiKiJ+XcWdRul7UxQJ6KmRushv1GjRiExMbEeqkJETdXVO5xFnYiaNq2T0jt06IBly5bh559/RmBgIKytlRNM582bp7PKEVHTwAR1ImrqtA6oPv74Y7Rs2RInT57EyZMnld6TSCQMqIhIRWRQB8xKOAWJpKJnCqgY9pvdv71e60VEpCtaB1RXr16tj3oQURMmS1BfnZqByzmFkEoFyqQCS785i1e+PANPe2tEBnVAiJ+LvqtKRFQrWudQKRJCQAhRc0EiavYUZ1F/ob8XAODvolIuS0NETUKtAqpPP/0UXbp0gaWlJSwtLdG1a1ds3bpV13UjoiYq9dxtpdeyP8tmJ5ziHFVE1ChpHVDFxcVh9uzZCA0NxY4dO7Bjxw6EhIRg1qxZeO+99+qjjkTUxKh76g+oCKzYW0VEjZHWOVRr1qzBBx98gClTpsi3Pf300+jcuTOWLl2KBQsW6LSCRNT0qHvqT0ZxjirmVBFRY6F1D1VmZib69Omjsr1Pnz7IzORflERUM3XL0igSAjiXmc/hPyJqNLQOqNq3b48dO3aobN++fTs6dOigk0oRUdOmuCxNFTEVAA7/EVHjofWQX2xsLMaPH4+DBw/iiSeeAAD8/PPPSE1NVRtoERGpE+LnghA/FySfzVSZo0pG9nL+tjQAaZxegYgMltY9VGPGjMGxY8dgb2+PxMREJCYmwt7eHsePH8fo0aPro45E1IQp9lZVpbhMyukViMigad1DBQCBgYFISEjQdV2IqJmS9VaFxB+sMlkdYMI6ERkurXuofvjhB+zZs0dl+549e5CUlKSTShFR81RTsjrARZWJyDBpHVAtWrQI5eXlKtuFEFi0aJFOKkVEzZPi8J+5iRHMTVRvUVxUmYgMkdZDfhkZGejUqZPK9o4dO+LSpUs6qRQRNV+y4T8AahPWBYD5g3z0V0EiIjW07qFq0aIFrly5orL90qVLsLbmX41EpDuKPVamxv+OA774BZeoISLDonVANXLkSERGRuLy5cvybZcuXcJLL72Ep59+WqeVIyKSLaq8ZmI3+bbScsEn/ojIoGgdUL377ruwtrZGx44d4enpCU9PTzz66KNo06YNVq5cWR91JCJC/L4MpUlAFeeo8o1JYo8VEemV1jlULVq0wJEjR5CSkoIzZ87A0tISXbt2Rb9+/eqjfkREACoWVFY3nUJxmRTAv7Oqb5jcnVMqEFGDq9U8VBKJBEOGDMGQIUN0XR8iIrWqW1AZ4BxVRKRfGg/5HT16FN99953Stk8//RSenp5wdHTE888/j+Li4lpVYt26dfDw8ICFhQV69eqF48ePV1m2tLQUy5Ytg7e3NywsLODv74/k5GSlMh4eHpBIJCr/5s6dW6v6EZH+aTpHFRdVJiJ90DigWrZsGf744w/5699//x0zZsxAUFAQFi1ahG+//RYrVqzQugLbt29HVFQUlixZglOnTsHf3x/BwcG4ffu22vIxMTHYuHEj1qxZg/T0dMyaNQujR4/G6dOn5WVOnDiBzMxM+b+UlBQAwLPPPqt1/YjIMGgyR5UME9aJqKFpHFClpaVh0KBB8tfbtm1Dr1698NFHHyEqKgrvv/9+rRZHjouLQ3h4OMLCwtCpUyds2LABVlZW2LRpk9ryW7duxauvvorQ0FB4eXlh9uzZCA0NxapVq+RlHBwc4OzsLP/33XffwdvbG/3799e6fkRkOGRP/F1YPhSrJwQAUN9jpTj8R0TUEDTOobp37x6cnJzkrw8cOIChQ4fKXz/22GO4ceOGVicvKSnByZMnER0dLd9mZGSEoKAgHD16VO0+xcXFsLCwUNpmaWmJw4cPV3mOhIQEREVFQVLFWEFxcbHScGVeXh6AiuHF0tJSra5Jtp/if6lu2J6605TacpCvPdZO8Mfa/ZdxPqtA5X0hgPOZ+fCJSYJnGyu8OMAbwZ2d1Byp9ppSexoCtqduNcf21Oe1ahxQOTk54erVq3B3d0dJSQlOnTqF2NhY+fv5+fkwNTXV6uR37txBeXm5UqAmO9f58+fV7hMcHIy4uDj069cP3t7eSE1Nxe7du9UuhwMAiYmJyM3NxbRp06qsx4oVK5SuRWbv3r2wsrLS/IIqkQ01km6wPXWnKbXlbE/gnTxj3CoCAOU/mgQESsqkuJCdj4htZzDdpxz+bapKa6+9ptSehoDtqVvNqT2Lior0dm6NA6rQ0FAsWrQI77zzDhITE2FlZYW+ffvK3//tt9/g7e1dL5VUtHr1aoSHh6Njx46QSCTw9vZGWFhYlUOEH3/8MYYOHQpXV9cqjxkdHY2oqCj567y8PLi7u2PIkCGws7PTuo6lpaVISUnB4MGDtQ4ySRXbU3eaalsat8tGxLYzSkvUVJDI/yuRAEfyWiD6uT46O29TbU99YXvqVnNsT9kIkz5oHFC98cYbeOaZZ9C/f3/Y2Nhgy5YtMDMzk7+/adMmradRsLe3h7GxMbKzs5W2Z2dnw9nZWe0+Dg4OSExMxMOHD3H37l24urpi0aJF8PLyUin7559/Yt++fdi9e3e19TA3N4e5ubnKdlNT0zp9COu6Pylje+pOU2vL4QFtYWJijNWpGbiSU4iSMqnK9ApCABeyCuAXuw+e9taIDOqgs+kVmlp76hvbU7eaU3vq8zo1Dqjs7e1x8OBB3L9/HzY2NjA2NlZ6f+fOnbCxsdHq5GZmZggMDERqaipGjRoFAJBKpUhNTUVERES1+1pYWMDNzQ2lpaXYtWsXxo0bp1Jm8+bNcHR0xLBhw7SqFxE1PoqLKofEH1Q7Z5VAxUSgnASUiHStVosjVw6mAKB169ZKPVaaioqKwkcffYQtW7bg3LlzmD17NgoLCxEWFgYAmDJlilLS+rFjx7B7925cuXIFhw4dQkhICKRSKRYuXKh0XKlUis2bN2Pq1KkwManV/KVE1EjVNGcVnwIkIl3TOqDStfHjx2PlypVYvHgxAgICkJaWhuTkZHmi+vXr15GZ+e9cMg8fPkRMTAw6deqE0aNHw83NDYcPH0bLli2Vjrtv3z5cv34d06dPb8jLISIDUHnOKnVxFScBJSJdMoium4iIiCqH+Pbv36/0un///khPT6/xmEOGDIEQun+ah4gaB02GAAGuAUhEuqH3HioiovpW3RCgQMWzgPO3pcE3Jok9VkRUKwyoiKjJUxwCVEeWrK6YsM6gioi0odGQ3zfffKPxAZ9++ulaV4aIqL7IhgCrG/4DlBPWOQRIRJrSKKCSTWlQE4lEUuWM5UREhiAyqANmJZxSMwnovxQT1nU5XxURNV0aDflJpVKN/jGYIiJDV/kJQHOTqm+DHP4jIk0ZxFN+REQNSfEJwOSzmVX2WHH4j4g0VauAqrCwEAcOHMD169dRUlKi9N68efN0UjEiooYg67FanZqBc5n5Ku8LAVzJKdRDzYioMdE6oDp9+jRCQ0NRVFSEwsJCtG7dGnfu3IGVlRUcHR0ZUBFRo1NTwroQgG9Mks7XACSipkPraRMWLFiAESNG4N69e7C0tMQvv/yCP//8E4GBgVi5cmV91JGIqEFUNV9VSTmnVCCi6mkdUKWlpeGll16CkZERjI2NUVxcDHd3d7z77rt49dVX66OOREQNonLCulmlhHWuAUhEVdE6oDI1NYWRUcVujo6OuH79OoCKRZNv3Lih29oRETWwED8XJM3vhwvLh9a4BuCeP7IbvH5EZJi0Dqi6deuGEydOAKhYV2/x4sX47LPPEBkZCT8/P51XkIhIXzztrdUGVUDFlAoR287gzN2qShBRc6J1QPXWW2/BxaUiIfPNN99Eq1atMHv2bOTk5GDjxo06ryARkb7UuAagBEj+iyt4EVEtnvLr0aOH/P8dHR2RnJys0woRERkKTaZUuP1ADxUjIoOj9Z9WAwcORG5ursr2vLw8DBw4UBd1IiIyGLKcqo7OtmqH/8oEMHztET75R9TMaR1Q7d+/X2UyTwB4+PAhDh06pJNKEREZmqqH/yS4mF3A6RSImjmNh/x+++03+f+np6cjKytL/rq8vBzJyclwc3PTbe2IiAyE4vDf+UzlyT+5RA0RaRxQBQQEQCKRQCKRqB3as7S0xJo1a3RaOSIiQyKbUd03JgnFZVKl97hEDVHzpnFAdfXqVQgh4OXlhePHj8PBwUH+npmZGRwdHWFsbFwvlSQiMiSe9tYqS9RIAHg5WOurSkSkZxoHVO3atQMASKXSGkoSETVtkUEdMCvhFCSSip4poGLYb/4gH73Wi4j0p1YTqFy+fBkvvvgigoKCEBQUhHnz5uHy5cu6rhsRkUGS5VP5OtnAWFIRUdmYm2BAR4ca9iSipkrrgGrPnj3o1KkTjh8/jq5du6Jr1644duwYOnfujJSUlPqoIxGRwQnxc8G3c/vgf73K4WRnjoLiMiSfzap5RyJqkrSe2HPRokVYsGAB3n77bZXt//3vfzF48GCdVY6IyNAZS4DxgW3x/k+X8dmx6xgZwKediZojrXuozp07hxkzZqhsnz59OtLT03VSKSKixuTZHm4wkgDHr/6NDq/9gJD4g5yTiqiZ0bqHysHBAWlpaejQoYPS9rS0NDg6OuqsYkREjcWZG/ch/Sc5vbRc4EJWPmYlnIKJkQTtHW0QGdSB81MRNXEa91AtW7YMRUVFCA8Px/PPP4933nkHhw4dwqFDh/D222/jhRdeQHh4eH3WlYjIIK356bLSsjSy6RTKpP8GV+yxImraNO6hio2NxaxZs/D666/D1tYWq1atQnR0NADA1dUVS5cuxbx58+qtokREhurq3SKlOakUcRZ1ouZB44BK/DPZikQiwYIFC7BgwQLk51esvm5ra1s/tSMiagQ821jhYnZB1UGVAM5n5sM3Jgme9tYcAiRqgrRKSpdUWhXU1taWwRQRNXsvDvCuYuHkfwkAxWVSDgESNVFaJaX7+PioBFWV/f3333WqEBFRYxPc2Um+cHJGdgHKpFX1Vf2bXzU74RR8nW3ZW0XURGgVUMXGxqJFixb1VRciokZLtnAyACSfzcTq1AxcySlESZlU7VCgAOS9VW4tLXCnoITDgUSNmFYB1YQJEzg1AhFRDRSDq5D4gyoLKcvItt3MfQiAARZRY6ZxDlVNQ31ERKQqMqhDjflVMooBlmK+VftXOVkokaHT+ik/IiLSnGwh5dWpGTifqb6nqirq5rNi7xWRYdK4h0oqlXK4j4ioFkL8XJA0vx8+mNwdgGa9VZVV1XvFXisiw6D1Wn5ERFQ7st6qjs62MDcxQttWlgBqH2DJJgwlIv3Tei0/IiKqPcWEdUD5iUAHW3P8de8BJJKKyUBrIgRwLjMfIfEHOfxHpGcMqIiI9KiqAKum+awUMb+KSP845EdEZEBk+VaX3grFhsnd8ahLzcODzK8i0j/2UBERGaiqeq/OZeZXux9nYydqeOyhIiJqJGS9Vx2dbaFJHrvibOzsrSKqXwyoiIgaGW0nC+XTgET1jwEVEVEjo+30C4pPA7Kniqh+MIeKiKgRqiq/qrrZ2Pk0IFH9YQ8VEVEToMls7HwakKj+6D2gWrduHTw8PGBhYYFevXrh+PHjVZYtLS3FsmXL4O3tDQsLC/j7+yM5OVml3M2bNzF58mS0adMGlpaW6NKlC3799df6vAwiIoOgOBxYE+ZXEemOXgOq7du3IyoqCkuWLMGpU6fg7++P4OBg3L59W235mJgYbNy4EWvWrEF6ejpmzZqF0aNH4/Tp0/Iy9+7dwxNPPAFTU1MkJSUhPT0dq1atQqtWrRrqsoiI9EqbpwGZX0WkG3oNqOLi4hAeHo6wsDB06tQJGzZsgJWVFTZt2qS2/NatW/Hqq68iNDQUXl5emD17NkJDQ7Fq1Sp5mXfeeQfu7u7YvHkzevbsCU9PTwwZMgTe3t4NdVlERAZBm6cBOfxHVDd6S0ovKSnByZMnER0dLd9mZGSEoKAgHD16VO0+xcXFsLCwUNpmaWmJw4cPy19/8803CA4OxrPPPosDBw7Azc0Nc+bMQXh4eJV1KS4uRnFxsfx1Xl4egIohxtLSUq2vTbZPbfYlVWxP3WFb6paht+cgX3usneCPtfsv48qdIjjYmOFm7kO1awXKAq/4fRcxyNdeL/U19PZsbJpje+rzWiVCaLIEp+7dunULbm5uOHLkCHr37i3fvnDhQhw4cADHjh1T2WfSpEk4c+YMEhMT4e3tjdTUVIwcORLl5eXygEgWcEVFReHZZ5/FiRMnMH/+fGzYsAFTp05VW5elS5ciNjZWZfvnn38OKysrXVwuEZFBOHNXguS/jHCrCIDaAUEBEwngaAmEtJXCv41eviKIaqWoqAiTJk3C/fv3YWdn16DnblQBVU5ODsLDw/Htt99CIpHA29sbQUFB2LRpEx48eAAAMDMzQ48ePXDkyBH5fvPmzcOJEyeq7fmq3EPl7u6OO3fu1OoHUlpaipSUFAwePBimpqZa70/K2J66w7bUrcbcnsPXHsHF7IIqp1iQMTGSwNvBGi8O8EZwZ6d6rVNjbk9D1BzbMy8vD/b29noJqPQ25Gdvbw9jY2NkZ2crbc/Ozoazs7PafRwcHJCYmIiHDx/i7t27cHV1xaJFi+Dl5SUv4+Ligk6dOint9+ijj2LXrl1V1sXc3Bzm5uYq201NTev0Iazr/qSM7ak7bEvdaoztuWCwD2YlnFI7/KeoTCpwMbsAEdvOYMPk7g0yZ1VjbE9D1pzaU5/XqbekdDMzMwQGBiI1NVW+TSqVIjU1VanHSh0LCwu4ubmhrKwMu3btwsiRI+XvPfHEE7hw4YJS+YsXL6Jdu3a6vQAiokas8mzr1eWty+Kt+dvS4BuTxCcCidTQ61N+UVFR+Oijj7BlyxacO3cOs2fPRmFhIcLCwgAAU6ZMUUpaP3bsGHbv3o0rV67g0KFDCAkJgVQqxcKFC+VlFixYgF9++QVvvfUWLl26hM8//xwffvgh5s6d2+DXR0RkyGTTK1xYPhS+GkyxUFwm5YSgRFXQa0A1fvx4rFy5EosXL0ZAQADS0tKQnJwMJ6eKcfrr168jM/PfX9iHDx8iJiYGnTp1wujRo+Hm5obDhw+jZcuW8jKPPfYYvvrqK3zxxRfw8/PDG2+8gfj4ePznP/9p6MsjImo0uOAyUd3ofS2/iIgIREREqH1v//79Sq/79++P9PT0Go85fPhwDB8+XBfVIyJqFmRDgKtTM5CRXYAyafXp6kIA5zPz4RuTxDUBiWAAARURERkGxQWXZYstX8kpBFAx3FeZ+Ge7bAjQxEiC9o42DK6oWdL7Wn5ERGR4FPOrVk8IAFD1cKCsL6tMKphfRc0WAyoiIqqWtk8EMr+KmiMO+RERUY0UhwND4g/iQlZ+lZOCCgH5UCFRc8EeKiIi0oqmTwRyzipqThhQERGRVhSHAE2M1EdVnLOKmhsO+RERkdaqeiJQKgRKy/8dDFTMqeKTf9SUsYeKiIjqRPGJQCM144BCAOcy8zn8R00aAyoiItIZT3vrKp8C5PAfNWUMqIiISGeqS1gXACTgIsvUNDGgIiIinVFMWFdHNrs6E9apqWFARUREOiXLqerobFvjJKAAMDvhFELiD2LPH9kNUT2iesGAioiI6oWm81UJVORXRWw7gzN3ayhMZKAYUBERUb2ovGSNuUnVXzmywCv5L34tUePEeaiIiKjeVJ6valbCKUgkFVMpVCYEcPtBA1eQSEf4pwARETUIxR6rqgb2ygQwfO0RJqpTo8OAioiIGowsYf2Dyd0BqMuvkuBidgGf/qNGhwEVERE1uOp6qzhfFTVGDKiIiEgvZL1VZmqS1TlfFTU2DKiIiEivqluuBlBeYJnIUDGgIiIivdJkviousEyGjtMmEBGRXsnyqeL3XcSl7HwYGxujuEyqtqxs+M+tpQXuFJTA094akUEd5FMzEOkLAyoiItK7ED8XDPK1xw8//ADjdgGI2HZG7XxVspc3cx8C+DfAMjGSoL2jDYMr0hsO+RERkUEJ7uxU7QLLimQBVplUMHmd9IoBFRERGRxNF1hWxOR10icGVEREZLA0XWBZRgjgfGY+56+iBseAioiIDFblBZbbtrIEUMMTgQDnr6IGx6R0IiIyaIoLLAMViyyvTs1ARnYByqRqVln+h+IQIBPVqb4xoCIiokZFMcCSBVdXcgpRUiZF5fBKCOBKTmHDV5KaHQ75ERFRoyVLXr+wfCh8q0hgLymTMp+K6h0DKiIiahKqSmAXAPOpqN4xoCIioiZBMYG9ck8Vp1Sg+saAioiImgzZEKCZierXG6dUoPrEgIqIiJocT3trtflUnFKB6gsDKiIianJqmhCUQ4CkawyoiIioyak8Iaja3ioBnMvM5/Af6QTnoSIioiZJcb6qkPiDuJCVrzJPFfDvE4BuLS1wp6AEnvbWiAzqwMlASSvsoSIioiavuiFAWZB1M/ch86uo1hhQERFRk6c4BFgT5ldRbTCgIiKiZkE2pYK6eaoqY34VaYsBFRERNSs1PQGoiMN/pCkGVERE1KxUfgKwbStLAFXnV3H4jzTBp/yIiKjZUXwCEACSz2ZidWoGzmXmq5QVAriSU9iQ1aNGiD1URETU7NWUXyUEuGQNVYsBFRER0T+qyq8qKZdySgWqlkEEVOvWrYOHhwcsLCzQq1cvHD9+vMqypaWlWLZsGby9vWFhYQF/f38kJycrlVm6dCkkEonSv44dO9b3ZRARUSNXOb+q8iLLzKmiqug9oNq+fTuioqKwZMkSnDp1Cv7+/ggODsbt27fVlo+JicHGjRuxZs0apKenY9asWRg9ejROnz6tVK5z587IzMyU/zt8+HBDXA4RETVysuG/C8uHcska0pjeA6q4uDiEh4cjLCwMnTp1woYNG2BlZYVNmzapLb9161a8+uqrCA0NhZeXF2bPno3Q0FCsWrVKqZyJiQmcnZ3l/+zt7RvicoiIqAnxtLeucs4q2fDfE2+nMr+K9PuUX0lJCU6ePIno6Gj5NiMjIwQFBeHo0aNq9ykuLoaFhYXSNktLS5UeqIyMDLi6usLCwgK9e/fGihUr8Mgjj1R5zOLiYvnrvLw8ABXDi6WlpVpfl2yf2uxLqtieusO21C22p24ZYntGPOWFiG1nIJFU9EwpUlyyBvg3wFo7wR/BnZ0atqJqGGJ71jd9XqtEiMofkYZz69YtuLm54ciRI+jdu7d8+8KFC3HgwAEcO3ZMZZ9JkybhzJkzSExMhLe3N1JTUzFy5EiUl5fLg6KkpCQUFBTA19cXmZmZiI2Nxc2bN3H27FnY2qouO7B06VLExsaqbP/8889hZWWlwysmIqLG5sxdCZL/MsKtIgA1zLEugYCLFfBf//IGqRspKyoqwqRJk3D//n3Y2dk16LkbXUCVk5OD8PBwfPvtt5BIJPD29kZQUBA2bdqEBw8eqD1Pbm4u2rVrh7i4OMyYMUPlfXU9VO7u7rhz506tfiClpaVISUnB4MGDYWpqqvX+pIztqTtsS91ie+qWobfn8LVHcDG7ADV9aZqZGOGPJUENUqfqGHp71oe8vDzY29vrJaDS65Cfvb09jI2NkZ2drbQ9Ozsbzs7OavdxcHBAYmIiHj58iLt378LV1RWLFi2Cl5dXledp2bIlfHx8cOnSJbXvm5ubw9zcXGW7qalpnT6Edd2flLE9dYdtqVtsT90y1PZcMNgHsxJOqR3+U1RaJsWIdUcRGdRBafJQfTHU9qwP+rxOvSalm5mZITAwEKmpqfJtUqkUqampSj1W6lhYWMDNzQ1lZWXYtWsXRo4cWWXZgoICXL58GS4u+v9gExFR46TpkjUCXAOwOdL70jNRUVGYOnUqevTogZ49eyI+Ph6FhYUICwsDAEyZMgVubm5YsWIFAODYsWO4efMmAgICcPPmTSxduhRSqRQLFy6UH/Pll1/GiBEj0K5dO9y6dQtLliyBsbExJk6cqJdrJCKipqGqJWvOZ+YrDQUqzldlCL1UVP/0HlCNHz8eOTk5WLx4MbKyshAQEIDk5GQ4OVU8IXH9+nUYGf3bkfbw4UPExMTgypUrsLGxQWhoKLZu3YqWLVvKy/z111+YOHEi7t69CwcHBzz55JP45Zdf4ODg0NCXR0RETZgswPKNSUJxmVTpPSGA85n58I1Jgqe9tcEMAVL90HtABQARERGIiIhQ+97+/fuVXvfv3x/p6enVHm/btm26qhoREVGNPO2tcSErXyVhXQBKS9ZsmNydQVUTpfeJPYmIiBq7qtYAlOGSNU0fAyoiIqI6qpywziVrmh8GVERERDqguAagr7NtjUvWMKhqWhhQERER6Vh1Q4Ac/muaGFARERHpmOIQoDqKTwByCLBpYEBFRERUD2RDgB2rGP6r/AQgg6rGjQEVERFRPdLkCUAAmJ1wir1VjRgDKiIionqkyROAAJesaewYUBEREdUzTZ8AZMJ648WAioiIqAHVOATIhPVGiQEVERFRA1IcAqyup4oJ642LQazlR0RE1JzIFlVOPpuJWQmnIJFU9ExVppiw7utsi/4+9jhw8Q6u3inkgssGhj1UREREeqJNwvr5rHxsPHgV57Py2XtlgBhQERER6ZGmCeuVMYHdsDCgIiIiMhA1JaxXJgRwJaewXutEmmFARUREZCA0SVivrKRMyqcBDQADKiIiIgMiGwL8YHJ3AP/2VnFCUMPGp/yIiIgMkKy3anVqBq7kFMLLwRr9OjjgYEYOzmfmQ/GhQNn/z9+WBiANnvbWiHjKq+Er3YwxoCIiIjJQsukVFEXjUfjGJKG4TKpSXrbtQlY+IradgRGMsf7KESwY7MPpFeoZAyoiIqJGxtPeGheylHupFMm2SyHBxewCzEo4BbeWFrhTUML5q+oJc6iIiIgaGW2eBpQFVzdzH3L+qnrEgIqIiKiRqTwhqLmJ5l/nirOv8+lA3eGQHxERUSOkmF9V0xI26ig+HcjhwLpjDxUREVEjp9hjZWKk6QxWHA7UJfZQERERNQGVe6zi913Epex8OLWwxM3chxr1XikuZ8NeKu2wh4qIiKiJCfFzwbdz+2DV4+XY/1I/rWZfFwI4l5nP/CotMaAiIiJq4qqafb06HP7TDgMqIiKiZqLy04FtW1kCUB9gKQ7/Uc2YQ0VERNSMVJ59PflsJlanZuBcZr5KWSGA85n58I1J4hOANWAPFRERUTMmGw6sKsdKAHwCUAMMqIiIiKjG2dc5BFg9BlRERESkkl+ltrdKAFdyChu8bo0BAyoiIiIC8O/w34XlQ+FbxRBgSZmUUyqowYCKiIiIVFQ1BKi4ZA2Dqn8xoCIiIiIVikOAlXuqmE+ligEVERERqSUbAjQzUQ0XFKdU4BAgAyoiIiKqgae9NadUqAEDKiIiIqqWJlMqAMDshFPNtreKARURERFVS5MpFYDmnbDOgIqIiIhqpMmUCkDzTVhnQEVERERaqXEIsBlOAMqAioiIiLRS3ZQKMlIhmtUTgAyoiIiISGuyIcAPJncHoNpbVVoumtUTgAyoiIiIqNYqJ6xXnrOqueRUmei7AkRERNS4hfi5IMTPBQDgG5Ok8n5zyKkyiB6qdevWwcPDAxYWFujVqxeOHz9eZdnS0lIsW7YM3t7esLCwgL+/P5KTk6ss//bbb0MikSAyMrIeak5ERESKqpoEtKkvqqz3gGr79u2IiorCkiVLcOrUKfj7+yM4OBi3b99WWz4mJgYbN27EmjVrkJ6ejlmzZmH06NE4ffq0StkTJ05g48aN6Nq1a31fBhEREaH5Lqqs9yG/uLg4hIeHIywsDACwYcMGfP/999i0aRMWLVqkUn7r1q147bXXEBoaCgCYPXs29u3bh1WrViEhIUFerqCgAP/5z3/w0UcfYfny5dXWobi4GMXFxfLXeXl5ACp6w0pLS7W+Jtk+tdmXVLE9dYdtqVtsT91ie+qWvtpzkK891k7wx9r9l3Ehq0A+izrwTz4VgPnb0iCQBs82VnhxgDeCOzvp5Nz6/OzoNaAqKSnByZMnER0dLd9mZGSEoKAgHD16VO0+xcXFsLCwUNpmaWmJw4cPK22bO3cuhg0bhqCgoBoDqhUrViA2NlZl+969e2FlZaXp5ahISUmp9b6kiu2pO2xL3WJ76hbbU7f01Z6zPYGXso1RJpS7qirW/ysHIMGF7HxEbDuD6T7l8G8j1B5HG0VFRXU+Rm3pNaC6c+cOysvL4eSkHJk6OTnh/PnzavcJDg5GXFwc+vXrB29vb6SmpmL37t0oLy+Xl9m2bRtOnTqFEydOaFSP6OhoREVFyV/n5eXB3d0dQ4YMgZ2dndbXVVpaipSUFAwePBimpqZa70/K2J66w7bULbanbrE9dcsQ2nP9lSO4mK3cS1VBIv+vRAIcyWuB6Of61Pl8shEmfdD7kJ+2Vq9ejfDwcHTs2BESiQTe3t4ICwvDpk2bAAA3btzA/PnzkZKSotKTVRVzc3OYm5urbDc1Na3Th7Cu+5MytqfusC11i+2pW2xP3dJney4Y7INZCacgkVQ86aeOEMDVO0U6qaM+Pzd6TUq3t7eHsbExsrOzlbZnZ2fD2dlZ7T4ODg5ITExEYWEh/vzzT5w/fx42Njbw8vICAJw8eRK3b99G9+7dYWJiAhMTExw4cADvv/8+TExMlHqyiIiIqP5UnqPK3EQ17JBIAC8Haz3UTrf02kNlZmaGwMBApKamYtSoUQAAqVSK1NRUREREVLuvhYUF3NzcUFpail27dmHcuHEAgEGDBuH3339XKhsWFoaOHTviv//9L4yNjevlWoiIiEiV4hxVyWczlXqsZP+dP8hHz7WsO70P+UVFRWHq1Kno0aMHevbsifj4eBQWFsqf+psyZQrc3NywYsUKAMCxY8dw8+ZNBAQE4ObNm1i6dCmkUikWLlwIALC1tYWfn5/SOaytrdGmTRuV7URERNRwZD1Wq1MzcCWnEF4O1pg/yAchfupHpRoTvQdU48ePR05ODhYvXoysrCwEBAQgOTlZnqh+/fp1GBn920X48OFDxMTE4MqVK7CxsUFoaCi2bt2Kli1b6ukKiIiISFOKPVZNid4DKgCIiIiocohv//79Sq/79++P9PR0rY5f+RhEREREuqT3mdKJiIiIGjsGVERERER1xICKiIiIqI4YUBERERHVEQMqIiIiojpiQEVERERURwyoiIiIiOqIARURERFRHTGgIiIiIqojBlREREREdWQQS88YGiEEACAvL69W+5eWlqKoqAh5eXkwNTXVZdWaJban7rAtdYvtqVtsT91qju0p+96WfY83JAZUauTn5wMA3N3d9VwTIiIi0lZ+fj5atGjRoOeUCH2EcQZOKpXi1q1bsLW1hUQi0Xr/vLw8uLu748aNG7Czs6uHGjYvbE/dYVvqFttTt9ieutUc21MIgfz8fLi6usLIqGGzmthDpYaRkRHatm1b5+PY2dk1mw9xQ2B76g7bUrfYnrrF9tSt5taeDd0zJcOkdCIiIqI6YkBFREREVEcMqOqBubk5lixZAnNzc31XpUlge+oO21K32J66xfbULbZnw2JSOhEREVEdsYeKiIiIqI4YUBERERHVEQMqIiIiojpiQEVERERURwyo1FixYgUee+wx2NrawtHREaNGjcKFCxeUyjx8+BBz585FmzZtYGNjgzFjxiA7O1v+/pkzZzBx4kS4u7vD0tISjz76KFavXq1yruLiYrz22mto164dzM3N4eHhgU2bNtX7NTakhmxPmZ9//hkmJiYICAior8vSm4Zqz927d2Pw4MFwcHCAnZ0devfujT179jTINTakhvx87t+/H927d4e5uTnat2+PTz75pL4vr8Hpoj0BYN68eQgMDIS5uXmVv8d79uzB448/DltbWzg4OGDMmDG4du1aPV2ZfjRkewohsHLlSvj4+MDc3Bxubm5488036+vSmhwGVGocOHAAc+fOxS+//IKUlBSUlpZiyJAhKCwslJdZsGABvv32W+zcuRMHDhzArVu38Mwzz8jfP3nyJBwdHZGQkIA//vgDr732GqKjo7F27Vqlc40bNw6pqan4+OOPceHCBXzxxRfw9fVtsGttCA3ZngCQm5uLKVOmYNCgQQ1yfQ2todrz4MGDGDx4MH744QecPHkSAwYMwIgRI3D69OkGvd761lDtefXqVQwbNgwDBgxAWloaIiMjMXPmzCYXpOqiPWWmT5+O8ePHqz3P1atXMXLkSAwcOBBpaWnYs2cP7ty5o/Y4jVlDtScAzJ8/H//3f/+HlStX4vz58/jmm2/Qs2fPermuJklQjW7fvi0AiAMHDgghhMjNzRWmpqZi586d8jLnzp0TAMTRo0erPM6cOXPEgAED5K+TkpJEixYtxN27d+uv8gaovtpTZvz48SImJkYsWbJE+Pv767z+hqa+21NRp06dRGxsrG4qbqDqqz0XLlwoOnfurFRm/PjxIjg4WMdXYFjq2p5V/R7v3LlTmJiYiPLycvm2b775RkgkElFSUqL7CzEQ9dWe6enpwsTERJw/f77e6t7UsYdKA/fv3wcAtG7dGkDFX6OlpaUICgqSl+nYsSMeeeQRHD16tNrjyI4BAN988w169OiBd999F25ubvDx8cHLL7+MBw8e1NOVGIb6ak8A2Lx5M65cuYIlS5bUQ80NU322pyKpVIr8/PxqyzQF9dWeR48eVToGAAQHB1d7jKZAV+1ZWWBgIIyMjLB582aUl5fj/v372Lp1K4KCgmBqaqrbizAg9dWe3377Lby8vPDdd9/B09MTHh4emDlzJv7++2/dXkATxsWRayCVShEZGYknnngCfn5+AICsrCyYmZmhZcuWSmWdnJyQlZWl9jhHjhzB9u3b8f3338u3XblyBYcPH4aFhQW++uor3LlzB3PmzMHdu3exefPmersmfarP9szIyMCiRYtw6NAhmJg0j492fbZnZStXrkRBQQHGjRuns/obmvpsz6ysLDg5OakcIy8vDw8ePIClpaVuL8YA6Ko91fH09MTevXsxbtw4vPDCCygvL0fv3r3xww8/6PISDEp9tueVK1fw559/YufOnfj0009RXl6OBQsWYOzYsfjxxx91eRlNVvP41qmDuXPn4uzZszh8+HCtj3H27FmMHDkSS5YswZAhQ+TbpVIpJBIJPvvsM/nq2HFxcRg7dizWr1/fJG+w9dWe5eXlmDRpEmJjY+Hj46Or6hq8+vx8Kvr8888RGxuLr7/+Go6OjrU+l6FrqPZsLnTRnlXJyspCeHg4pk6diokTJyI/Px+LFy/G2LFjkZKSAolEovNz6lt9tqdUKkVxcTE+/fRT+T30448/RmBgIC5cuNDkcnvrA4f8qhEREYHvvvsOP/30E9q2bSvf7uzsjJKSEuTm5iqVz87OhrOzs9K29PR0DBo0CM8//zxiYmKU3nNxcYGbm5s8mAKARx99FEII/PXXX7q/ID2rz/bMz8/Hr7/+ioiICJiYmMDExATLli3DmTNnYGJi0iT/wqrvz6fMtm3bMHPmTOzYsUNlyKopqe/2dHZ2VnnyKjs7G3Z2dk3yjyddtGd11q1bhxYtWuDdd99Ft27d0K9fPyQkJCA1NRXHjh3T1WUYjPpuTxcXF5iYmCj9Qfroo48CAK5fv163yjcX+k7iMkRSqVTMnTtXuLq6iosXL6q8L0sC/PLLL+Xbzp8/r5IEePbsWeHo6CheeeUVtefZuHGjsLS0FPn5+fJtiYmJwsjISBQVFenwivSrIdqzvLxc/P7770r/Zs+eLXx9fcXvv/8uCgoK6ufi9KChPp9CCPH5558LCwsLkZiYqNuLMCAN1Z4LFy4Ufn5+StsmTpzY5JLSddWeMlUlUUdFRYmePXsqbbt165YAIH7++ee6X4iBaKj23LNnjwAgLl26JN+WlpYmAIgLFy7o5mKaOAZUasyePVu0aNFC7N+/X2RmZsr/KQY5s2bNEo888oj48ccfxa+//ip69+4tevfuLX//999/Fw4ODmLy5MlKx7h9+7a8TH5+vmjbtq0YO3as+OOPP8SBAwdEhw4dxMyZMxv0eutbQ7VnZU31Kb+Gas/PPvtMmJiYiHXr1imVyc3NbdDrrW8N1Z5XrlwRVlZW4pVXXhHnzp0T69atE8bGxiI5OblBr7e+6aI9hRAiIyNDnD59WrzwwgvCx8dHnD59Wpw+fVoUFxcLIYRITU0VEolExMbGiosXL4qTJ0+K4OBg0a5duyb1B2lDtWd5ebno3r276Nevnzh16pT49ddfRa9evcTgwYMb9HobMwZUagBQ+2/z5s3yMg8ePBBz5swRrVq1ElZWVmL06NEiMzNT/v6SJUvUHqNdu3ZK5zp37pwICgoSlpaWom3btiIqKqpJ3QyEaNj2VNRUA6qGas/+/furLTN16tSGu9gG0JCfz59++kkEBAQIMzMz4eXlpXSOpkIX7SlE1Z+/q1evyst88cUXolu3bsLa2lo4ODiIp59+Wpw7d66BrrRhNGR73rx5UzzzzDPCxsZGODk5iWnTpjW7aX3qQiKEENoPFBIRERGRDJPSiYiIiOqIARURERFRHTGgIiIiIqojBlREREREdcSAioiIiKiOGFARERER1REDKiIiIqI6YkBFREREVEcMqIiIiIjqiAEVETVK06ZNg0QigUQigampKZycnDB48GBs2rQJUqlU4+N88sknaNmyZf1VlIiaBQZURNRohYSEIDMzE9euXUNSUhIGDBiA+fPnY/jw4SgrK9N39YioGWFARUSNlrm5OZydneHm5obu3bvj1Vdfxddff42kpCR88sknAIC4uDh06dIF1tbWcHd3x5w5c1BQUAAA2L9/P8LCwnD//n15b9fSpUsBAMXFxXj55Zfh5uYGa2tr9OrVC/v379fPhRKRwWNARURNysCBA+Hv74/du3cDAIyMjPD+++/jjz/+wJYtW/Djjz9i4cKFAIA+ffogPj4ednZ2yMzMRGZmJl5++WUAQEREBI4ePYpt27bht99+w7PPPouQkBBkZGTo7dqIyHBJhBBC35UgItLWtGnTkJubi8TERJX3JkyYgN9++w3p6ekq73355ZeYNWsW7ty5A6AihyoyMhK5ubnyMtevX4eXlxeuX78OV1dX+fagoCD07NkTb731ls6vh4gaNxN9V4CISNeEEJBIJACAffv2YcWKFTh//jzy8vJQVlaGhw8foqioCFZWVmr3//3331FeXg4fHx+l7cXFxWjTpk2915+IGh8GVETU5Jw7dw6enp64du0ahg8fjtmzZ+PNN99E69atcfjwYcyYMQMlJSVVBlQFBQUwNjbGyZMnYWxsrPSejY1NQ1wCETUyDKiIqEn58ccf8fvvv2PBggU4efIkpFIpVq1aBSOjipTRHTt2KJU3MzNDeXm50rZu3bqhvLwct2/fRt++fRus7kTUeDGgIqJGq7i4GFlZWSgvL0d2djaSk5OxYsUKDB8+HFOmTMHZs2dRWlqKNWvWYMSIEfj555+xYcMGpWN4eHigoKAAqamp8Pf3h5WVFXx8fPCf//wHU6ZMwapVq9CtWzfk5OQgNTUVXbt2xbBhw/R0xURkqPiUHxE1WsnJyXBxcYGHhwdCQkLw008/4f3338fXX38NY2Nj+Pv7Iy4uDu+88w78/Pzw2WefYcWKFUrH6NOnD2bNmoXx48fDwcEB7777LgBg8+bNmDJlCl566SX4+vpi1KhROHHiBB555BF9XCoRGTg+5UdERERUR+yhIiIiIqojBlREREREdcSAioiIiKiOGFARERER1REDKiIiIqI6YkBFREREVEcMqIiIiIjqiAEVERERUR0xoCIiIiKqIwZURERERHXEgIqIiIiojv4f39GK/G6c6OEAAAAASUVORK5CYII=", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "score_df = pd.DataFrame(score_history_c)\n", + "score_df = score_df.sort_values('date')\n", + "\n", + "plt.plot(score_df['date'], score_df['total_score'], marker='o',markersize=4)\n", + "plt.gca().invert_xaxis() # remonter le temps\n", + "plt.title(\"Evolution de la Somme des Scores avec pénalité proportionnelle\")\n", + "plt.xlabel(\"Date\")\n", + "plt.ylabel(\"Total Score\")\n", + "plt.grid(True)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "eb308001-cdec-43b1-935d-0dc9dc59c192", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
datetotal_score
1292015-01-310.936891
1282015-02-280.937214
1272015-03-310.937968
1262015-04-300.938711
1252015-05-310.938969
.........
42025-06-300.998601
32025-07-310.998947
22025-08-310.999188
12025-09-300.999448
02025-10-311.000000
\n", + "

130 rows × 2 columns

\n", + "
" + ], + "text/plain": [ + " date total_score\n", + "129 2015-01-31 0.936891\n", + "128 2015-02-28 0.937214\n", + "127 2015-03-31 0.937968\n", + "126 2015-04-30 0.938711\n", + "125 2015-05-31 0.938969\n", + ".. ... ...\n", + "4 2025-06-30 0.998601\n", + "3 2025-07-31 0.998947\n", + "2 2025-08-31 0.999188\n", + "1 2025-09-30 0.999448\n", + "0 2025-10-31 1.000000\n", + "\n", + "[130 rows x 2 columns]" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "score_df" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "1579c56d-df62-4161-a0a0-89b35f197ca3", + "metadata": {}, + "outputs": [], + "source": [ + "def score_decay_vector_max(account, date_t, date_t1, tol=0.05, alpha=0.01):\n", + " \"\"\"\n", + " Score décroît proportionnellement à l'erreur relative,\n", + " en utilisant le max entre stock actuel et stock ajusté par flux comme dénominateur.\n", + " \"\"\"\n", + " # Récupérer tous les ISIN pour ce compte\n", + " isins_t = stocks_agg.loc[date_t, account].index if (date_t, account) in stocks_agg.index else []\n", + " isins_t1 = stocks_agg.loc[date_t1, account].index if (date_t1, account) in stocks_agg.index else []\n", + " isins_fl = flows_agg.loc[date_t, account].index if (date_t, account) in flows_agg.index else []\n", + " \n", + " all_isins = set(isins_t) | set(isins_t1) | set(isins_fl)\n", + " if len(all_isins) == 0:\n", + " return 1.0 # pas de données → pas de dégradation\n", + " \n", + " # Séries float complètes\n", + " st_t = pd.Series({isin: stocks_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " st_t1 = pd.Series({isin: stocks_agg.get((date_t1, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " fl_t = pd.Series({isin: flows_agg.get((date_t, account, isin), 0) for isin in all_isins}).astype(float).fillna(0)\n", + " \n", + " diff = np.abs(st_t.values - (st_t1.values + fl_t.values))\n", + " \n", + " # Dénominateur stable : max entre stock actuel et stock ajusté par flux\n", + " denom = np.maximum(st_t.values, st_t1.values + fl_t.values)\n", + " \n", + " # Calcul erreur relative\n", + " relative_error = np.zeros_like(diff)\n", + " nonzero_mask = denom != 0\n", + " relative_error[nonzero_mask] = diff[nonzero_mask] / denom[nonzero_mask]\n", + " \n", + " relative_error_clip = np.maximum(relative_error - tol, 0)\n", + " \n", + " decay_factor = max(0, 1 - alpha * np.nanmean(relative_error_clip))\n", + " return decay_factor" + ] + }, + { + "cell_type": "code", + "execution_count": 140, + "id": "e3c20686-38b7-473b-86e1-c8db01251098", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Analyse pour différentes sensibilités\n", + "\n", + "import matplotlib.pyplot as plt\n", + "score_df3 = pd.DataFrame(score_history_c)\n", + "score_df3 = score_df3.sort_values('date')\n", + "score_df2 = pd.DataFrame(score_history_c)\n", + "score_df2 = score_df2.sort_values('date')\n", + "score_df1 = pd.DataFrame(score_history)\n", + "score_df1 = score_df1.sort_values('date')\n", + "\n", + "plt.figure(figsize=(10,5))\n", + "\n", + "# Tracer chaque score_history\n", + "\n", + "plt.plot(score_df2['date'], score_df2['total_score'], marker='o',markersize=2, label='alpha=0.01')\n", + "plt.plot(score_df1['date'], score_df1['total_score'], marker='o',markersize=2, label='alpha=0.05')\n", + "plt.plot(score_df3['date'], score_df3['total_score'], marker='o',markersize=2, label='alpha=0.1')\n", + "\n", + "# Inverser l'axe X si tu veux remonter dans le temps\n", + "plt.gca().invert_xaxis()\n", + "\n", + "# Limites et labels\n", + "plt.ylim(0.9, 1.05)\n", + "plt.xlabel(\"Date\")\n", + "plt.ylabel(\"Total Score\")\n", + "plt.title(\"Évolution du score pour différentes sensibilités\")\n", + "\n", + "# Grille et légende\n", + "plt.grid(True)\n", + "plt.legend()\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 77, + "id": "d85c7835-b4f3-4d0e-a9e3-53bb3a1420c7", + "metadata": {}, + "outputs": [], + "source": [ + "score_history = []\n", + "score_history_accounts = []\n", + "\n", + "scores = aum_account.set_index('Registrar Account - ID')['weight'].to_dict()\n", + "\n", + "score_history.append({\n", + " \"date\": ref_date,\n", + " \"total_score\": sum(scores.values())\n", + "})\n", + "\n", + "for account, score in scores.items():\n", + " score_history_accounts.append({\n", + " \"date\": ref_date,\n", + " \"account\": account,\n", + " \"score\": score\n", + " })\n", + "\n", + "dates_sorted = sorted(df['Centralisation Date'].unique(), reverse=True)\n", + "\n", + "alpha = 0.01\n", + "tol = 0.05\n", + "\n", + "for i in range(1, len(dates_sorted)):\n", + "\n", + " date_t = dates_sorted[i-1]\n", + " date_t1 = dates_sorted[i]\n", + "\n", + " new_scores = scores.copy()\n", + "\n", + " for account in scores.keys():\n", + " decay = score_decay_vector_max(account, date_t, date_t1, tol=tol, alpha=alpha)\n", + " new_scores[account] = scores[account] * decay\n", + "\n", + " scores = new_scores\n", + "\n", + " # score total\n", + " score_history.append({\n", + " \"date\": date_t1,\n", + " \"total_score\": sum(scores.values())\n", + " })\n", + "\n", + " # score par compte\n", + " for account, score in scores.items():\n", + " score_history_accounts.append({\n", + " \"date\": date_t1,\n", + " \"account\": account,\n", + " \"score\": score\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": 78, + "id": "76c1d97f-3e73-4305-a161-4a12abc9ffc2", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
dateaccountscore
02025-10-314203500.048394
12025-10-313647650.041236
22025-10-312001274540.026188
32025-10-313129330.024981
42025-10-312001278090.024872
\n", + "
" + ], + "text/plain": [ + " date account score\n", + "0 2025-10-31 420350 0.048394\n", + "1 2025-10-31 364765 0.041236\n", + "2 2025-10-31 200127454 0.026188\n", + "3 2025-10-31 312933 0.024981\n", + "4 2025-10-31 200127809 0.024872" + ] + }, + "execution_count": 78, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "score_accounts_df = pd.DataFrame(score_history_accounts)\n", + "score_accounts_df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 126, + "id": "0e0239f2-ec24-4a43-925b-365a087de3e4", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "account_id = '200127375'\n", + "\n", + "df_account = (\n", + " score_accounts_df[score_accounts_df[\"account\"] == account_id]\n", + " .sort_values(\"date\")\n", + ")\n", + "\n", + "if len(df_account) == 0:\n", + " print(\"Compte non trouvé\")\n", + "else:\n", + " plt.figure(figsize=(8,4))\n", + " plt.plot(df_account[\"date\"], df_account[\"score\"], marker='o', markersize=4)\n", + "\n", + " plt.gca().invert_xaxis()\n", + " \n", + " plt.grid(True)\n", + "\n", + " plt.xlabel(\"Date\")\n", + " plt.ylabel(\"Score\")\n", + " plt.title(f\"Score du compte {account_id}\")\n", + "\n", + " plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 97, + "id": "dd2f09f1-e586-4f8d-925b-b7b8d2809541", + "metadata": {}, + "outputs": [], + "source": [ + "def evolution_compte(df, id1, id = 'Registrar Account - ID'):\n", + " def prepare_df(idt):\n", + " df_id = df[df[id] == idt].copy()\n", + " df_id['Centralisation Date'] = pd.to_datetime(df_id['Centralisation Date'])\n", + " df_agg = (\n", + " df_id\n", + " .groupby('Centralisation Date')['Quantity - AUM']\n", + " .sum()\n", + " .reset_index()\n", + " .sort_values('Centralisation Date')\n", + " )\n", + " return df_agg\n", + "\n", + " df1 = prepare_df(id1)\n", + "\n", + " plt.figure(figsize=(12, 6))\n", + "\n", + " # Courbe du compte\n", + " plt.plot(df1['Centralisation Date'], df1['Quantity - AUM'],\n", + " marker='.', linestyle='-', label=f'Account {id1}')\n", + "\n", + " plt.title(f\"Évolution AUM pour le compte {id1}\")\n", + " plt.xlabel(\"Date\")\n", + " plt.ylabel(\"Quantity - AUM\")\n", + " plt.grid(True)\n", + " plt.legend()\n", + " plt.tight_layout()\n", + " plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 103, + "id": "3a6d1927-200e-48ee-8e96-f09e182ae340", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "evolution_compte(df, '200127495') \n" + ] + }, + { + "cell_type": "code", + "execution_count": 99, + "id": "f13d2f9b-e4ce-4c1c-aa2b-55fb63b52616", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "evolution_compte(df, '200127479') " + ] + }, + { + "cell_type": "code", + "execution_count": 80, + "id": "bb58a3a7-34c8-4b1b-a31d-7861f9336725", + "metadata": {}, + "outputs": [], + "source": [ + "# Score : chute\n", + "\n", + "score_accounts_df = score_accounts_df.sort_values([\"account\",\"date\"])\n", + "\n", + "score_accounts_df[\"score_prev\"] = score_accounts_df.groupby(\"account\")[\"score\"].shift(1)\n", + "\n", + "score_accounts_df[\"relative_drop\"] = (\n", + " abs((score_accounts_df[\"score_prev\"] - score_accounts_df[\"score\"]) /\n", + " score_accounts_df[\"score\"]\n", + "))\n" + ] + }, + { + "cell_type": "code", + "execution_count": 120, + "id": "4415773b-c5e8-44fb-965d-e53f74037e2c", + "metadata": {}, + "outputs": [], + "source": [ + "ruptures = score_accounts_df[score_accounts_df[\"relative_drop\"] > 0.01]" + ] + }, + { + "cell_type": "code", + "execution_count": 121, + "id": "38ad7718-e561-4457-8bcd-a67e09f027d8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
dateaccountscorescore_prevrelative_drop
198232022-01-312001271750.0001440.0001430.010900
189612022-03-312001271750.0001470.0001440.016491
169272022-07-312001271830.0015600.0015400.012848
93852024-01-312001273310.0002450.0002140.127564
132802023-04-302001273450.0001880.0000001.000000
94012024-01-312001273450.0002160.0001880.129203
208352021-10-312001273560.0011550.0011330.019231
44572024-12-312001273560.0012010.0011820.015673
117852023-07-312001273750.0011770.0011640.011002
203842021-11-302001274430.0013830.0013680.010832
190912022-02-282001274430.0014470.0014010.031917
3982025-10-312001274550.0001710.0001690.011677
166212022-08-312001274790.0004700.0000001.000000
157592022-10-312001274790.0004750.0004700.010020
111572023-09-302001274950.0001890.0000001.000000
1662025-10-312001275540.0009410.0008500.096969
126142023-05-312001275790.0016000.0015680.019538
96822023-12-312001276130.0006790.0004700.307807
141322023-02-282001276390.0002390.0002340.020886
173992022-06-302001277310.0009730.0008240.152497
100342023-11-302001277430.0015490.0015210.017847
98062023-12-312001284730.0002710.0000001.000000
175312022-06-302001316480.0003490.0003410.021279
161792022-09-302001317360.0005210.0004860.066359
151862022-11-302001374670.0016310.0015800.031048
143242023-01-312001374670.0016560.0016370.011322
130312023-04-302001374670.0017420.0016670.043261
126002023-05-312001374670.0017660.0017420.013685
104452023-10-312001374670.0019890.0017780.105985
91522024-01-312001374670.0020190.0019980.010172
502025-10-312001379970.0046740.0046180.012157
205392021-11-303653040.0000460.0000450.012108
201082021-12-313653040.0000460.0000460.010058
196772022-01-313653040.0000480.0000460.039179
192462022-02-283653040.0000510.0000480.058795
175222022-06-303653040.0000530.0000520.011141
136432023-03-313653040.0000650.0000540.163563
132122023-04-303653040.0000660.0000650.015525
127812023-05-313653040.0000820.0000660.199242
119192023-07-313653040.0000870.0000830.048745
114882023-08-313653040.0000880.0000870.013252
89022024-02-293653040.0000920.0000910.012791
80402024-04-303653040.0000930.0000920.010325
76092024-05-313653040.0000950.0000930.012915
67472024-07-313653040.0000990.0000950.038845
63162024-08-313653040.0001000.0000990.011412
58852024-09-303653040.0001050.0001000.041680
50232024-11-303653040.0001110.0001050.046012
45922024-12-313653040.0001150.0001110.037865
24372025-05-313653040.0001390.0001180.147957
20062025-06-303653040.0003610.0001390.614563
15752025-07-313653040.0003710.0003610.027594
7132025-09-303653040.0003790.0003730.015225
\n", + "
" + ], + "text/plain": [ + " date account score score_prev relative_drop\n", + "19823 2022-01-31 200127175 0.000144 0.000143 0.010900\n", + "18961 2022-03-31 200127175 0.000147 0.000144 0.016491\n", + "16927 2022-07-31 200127183 0.001560 0.001540 0.012848\n", + "9385 2024-01-31 200127331 0.000245 0.000214 0.127564\n", + "13280 2023-04-30 200127345 0.000188 0.000000 1.000000\n", + "9401 2024-01-31 200127345 0.000216 0.000188 0.129203\n", + "20835 2021-10-31 200127356 0.001155 0.001133 0.019231\n", + "4457 2024-12-31 200127356 0.001201 0.001182 0.015673\n", + "11785 2023-07-31 200127375 0.001177 0.001164 0.011002\n", + "20384 2021-11-30 200127443 0.001383 0.001368 0.010832\n", + "19091 2022-02-28 200127443 0.001447 0.001401 0.031917\n", + "398 2025-10-31 200127455 0.000171 0.000169 0.011677\n", + "16621 2022-08-31 200127479 0.000470 0.000000 1.000000\n", + "15759 2022-10-31 200127479 0.000475 0.000470 0.010020\n", + "11157 2023-09-30 200127495 0.000189 0.000000 1.000000\n", + "166 2025-10-31 200127554 0.000941 0.000850 0.096969\n", + "12614 2023-05-31 200127579 0.001600 0.001568 0.019538\n", + "9682 2023-12-31 200127613 0.000679 0.000470 0.307807\n", + "14132 2023-02-28 200127639 0.000239 0.000234 0.020886\n", + "17399 2022-06-30 200127731 0.000973 0.000824 0.152497\n", + "10034 2023-11-30 200127743 0.001549 0.001521 0.017847\n", + "9806 2023-12-31 200128473 0.000271 0.000000 1.000000\n", + "17531 2022-06-30 200131648 0.000349 0.000341 0.021279\n", + "16179 2022-09-30 200131736 0.000521 0.000486 0.066359\n", + "15186 2022-11-30 200137467 0.001631 0.001580 0.031048\n", + "14324 2023-01-31 200137467 0.001656 0.001637 0.011322\n", + "13031 2023-04-30 200137467 0.001742 0.001667 0.043261\n", + "12600 2023-05-31 200137467 0.001766 0.001742 0.013685\n", + "10445 2023-10-31 200137467 0.001989 0.001778 0.105985\n", + "9152 2024-01-31 200137467 0.002019 0.001998 0.010172\n", + "50 2025-10-31 200137997 0.004674 0.004618 0.012157\n", + "20539 2021-11-30 365304 0.000046 0.000045 0.012108\n", + "20108 2021-12-31 365304 0.000046 0.000046 0.010058\n", + "19677 2022-01-31 365304 0.000048 0.000046 0.039179\n", + "19246 2022-02-28 365304 0.000051 0.000048 0.058795\n", + "17522 2022-06-30 365304 0.000053 0.000052 0.011141\n", + "13643 2023-03-31 365304 0.000065 0.000054 0.163563\n", + "13212 2023-04-30 365304 0.000066 0.000065 0.015525\n", + "12781 2023-05-31 365304 0.000082 0.000066 0.199242\n", + "11919 2023-07-31 365304 0.000087 0.000083 0.048745\n", + "11488 2023-08-31 365304 0.000088 0.000087 0.013252\n", + "8902 2024-02-29 365304 0.000092 0.000091 0.012791\n", + "8040 2024-04-30 365304 0.000093 0.000092 0.010325\n", + "7609 2024-05-31 365304 0.000095 0.000093 0.012915\n", + "6747 2024-07-31 365304 0.000099 0.000095 0.038845\n", + "6316 2024-08-31 365304 0.000100 0.000099 0.011412\n", + "5885 2024-09-30 365304 0.000105 0.000100 0.041680\n", + "5023 2024-11-30 365304 0.000111 0.000105 0.046012\n", + "4592 2024-12-31 365304 0.000115 0.000111 0.037865\n", + "2437 2025-05-31 365304 0.000139 0.000118 0.147957\n", + "2006 2025-06-30 365304 0.000361 0.000139 0.614563\n", + "1575 2025-07-31 365304 0.000371 0.000361 0.027594\n", + "713 2025-09-30 365304 0.000379 0.000373 0.015225" + ] + }, + "execution_count": 121, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ruptures" + ] + }, + { + "cell_type": "code", + "execution_count": 122, + "id": "3cecbc48-9ee8-4f81-b523-00561be292dc", + "metadata": {}, + "outputs": [], + "source": [ + "ruptures_recent = (\n", + " ruptures.sort_values(\"date\", ascending=False)\n", + " .groupby(\"account\")\n", + " .first()\n", + " .reset_index()\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 123, + "id": "30d11762-99be-4139-8b94-64f009290fdd", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
accountdatescorescore_prevrelative_drop
02001271752022-03-310.0001470.0001440.016491
12001271832022-07-310.0015600.0015400.012848
22001273312024-01-310.0002450.0002140.127564
32001273452024-01-310.0002160.0001880.129203
42001273562024-12-310.0012010.0011820.015673
52001273752023-07-310.0011770.0011640.011002
62001274432022-02-280.0014470.0014010.031917
72001274552025-10-310.0001710.0001690.011677
82001274792022-10-310.0004750.0004700.010020
92001274952023-09-300.0001890.0000001.000000
102001275542025-10-310.0009410.0008500.096969
112001275792023-05-310.0016000.0015680.019538
122001276132023-12-310.0006790.0004700.307807
132001276392023-02-280.0002390.0002340.020886
142001277312022-06-300.0009730.0008240.152497
152001277432023-11-300.0015490.0015210.017847
162001284732023-12-310.0002710.0000001.000000
172001316482022-06-300.0003490.0003410.021279
182001317362022-09-300.0005210.0004860.066359
192001374672024-01-310.0020190.0019980.010172
202001379972025-10-310.0046740.0046180.012157
213653042025-09-300.0003790.0003730.015225
\n", + "
" + ], + "text/plain": [ + " account date score score_prev relative_drop\n", + "0 200127175 2022-03-31 0.000147 0.000144 0.016491\n", + "1 200127183 2022-07-31 0.001560 0.001540 0.012848\n", + "2 200127331 2024-01-31 0.000245 0.000214 0.127564\n", + "3 200127345 2024-01-31 0.000216 0.000188 0.129203\n", + "4 200127356 2024-12-31 0.001201 0.001182 0.015673\n", + "5 200127375 2023-07-31 0.001177 0.001164 0.011002\n", + "6 200127443 2022-02-28 0.001447 0.001401 0.031917\n", + "7 200127455 2025-10-31 0.000171 0.000169 0.011677\n", + "8 200127479 2022-10-31 0.000475 0.000470 0.010020\n", + "9 200127495 2023-09-30 0.000189 0.000000 1.000000\n", + "10 200127554 2025-10-31 0.000941 0.000850 0.096969\n", + "11 200127579 2023-05-31 0.001600 0.001568 0.019538\n", + "12 200127613 2023-12-31 0.000679 0.000470 0.307807\n", + "13 200127639 2023-02-28 0.000239 0.000234 0.020886\n", + "14 200127731 2022-06-30 0.000973 0.000824 0.152497\n", + "15 200127743 2023-11-30 0.001549 0.001521 0.017847\n", + "16 200128473 2023-12-31 0.000271 0.000000 1.000000\n", + "17 200131648 2022-06-30 0.000349 0.000341 0.021279\n", + "18 200131736 2022-09-30 0.000521 0.000486 0.066359\n", + "19 200137467 2024-01-31 0.002019 0.001998 0.010172\n", + "20 200137997 2025-10-31 0.004674 0.004618 0.012157\n", + "21 365304 2025-09-30 0.000379 0.000373 0.015225" + ] + }, + "execution_count": 123, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ruptures_recent" + ] + }, + { + "cell_type": "code", + "execution_count": 124, + "id": "3da97871-c9e3-427b-9c8a-28034fe93cc2", + "metadata": {}, + "outputs": [], + "source": [ + "ruptures_recent = ruptures_recent.sort_values(\"relative_drop\", ascending=False)" + ] + }, + { + "cell_type": "code", + "execution_count": 125, + "id": "fed34a3d-c499-412b-83ee-690f22d3b923", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "53" + ] + }, + "execution_count": 125, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(ruptures)" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "72534b99-1125-43cf-a06c-f6958164cc2a", + "metadata": {}, + "outputs": [], + "source": [ + "# Pivot multi-index pour un accès rapide\n", + "stocks_dict = {}\n", + "for (date, account, isin), value in stocks_agg.items():\n", + " stocks_dict.setdefault(date, {}).setdefault(account, {})[isin] = value\n", + "\n", + "flows_dict = {}\n", + "for (date, account, isin), value in flows_agg.items():\n", + " flows_dict.setdefault(date, {}).setdefault(account, {})[isin] = value" + ] + }, + { + "cell_type": "code", + "execution_count": 131, + "id": "75c53c57-64ad-4c7a-829a-c77b4afdcc23", + "metadata": {}, + "outputs": [], + "source": [ + "# Fonction de chirurgie pour un compte avec rupture\n", + "\n", + "def surgery_for_account(account, date_t, date_t1, tol=0.01, alpha=0.01):\n", + " \"\"\"\n", + " Pour un compte donné et une rupture, trouver le code à date_t1\n", + " qui minimise la perte de score.\n", + " \"\"\"\n", + " st_t = stocks_dict.get(date_t, {}).get(account, {})\n", + " fl_t = flows_dict.get(date_t, {}).get(account, {})\n", + " \n", + " # Tous les ISIN impliqués\n", + " all_isins = set(st_t.keys())\n", + " \n", + " # Décroissance initiale avec le même code\n", + " st_t1_current = stocks_dict.get(date_t1, {}).get(account, {})\n", + " st_t_s = pd.Series({isin: st_t.get(isin,0) for isin in all_isins})\n", + " st_t1_s = pd.Series({isin: st_t1_current.get(isin,0) for isin in all_isins})\n", + " fl_t_s = pd.Series({isin: fl_t.get(isin,0) for isin in all_isins})\n", + " \n", + " diff = abs(st_t_s - (st_t1_s + fl_t_s))\n", + " denom = np.maximum(st_t_s, st_t1_s + fl_t_s)\n", + " relative_error = np.zeros_like(diff)\n", + " nonzero_mask = denom != 0\n", + " relative_error[nonzero_mask] = diff[nonzero_mask] / denom[nonzero_mask]\n", + " relative_error_clip = np.maximum(relative_error - tol, 0)\n", + " best_decay = max(0, 1 - alpha * np.nanmean(relative_error_clip))\n", + " best_code = account\n", + " \n", + " # Si rupture > tol, tester tous les codes à t1\n", + " if best_decay < 1 - tol:\n", + " for candidate in stocks_dict.get(date_t1, {}).keys():\n", + " st_c = stocks_dict[date_t1][candidate]\n", + " st_c_s = pd.Series({isin: st_c.get(isin,0) for isin in all_isins})\n", + " diff_c = abs(st_t_s - (st_c_s + fl_t_s))\n", + " denom_c = np.maximum(st_t_s, st_c_s + fl_t_s)\n", + " relative_error_c = np.zeros_like(diff_c)\n", + " nonzero_mask_c = denom_c != 0\n", + " relative_error_c[nonzero_mask_c] = diff_c[nonzero_mask_c] / denom_c[nonzero_mask_c]\n", + " relative_error_clip_c = np.maximum(relative_error_c - tol, 0)\n", + " decay_c = max(0, 1 - alpha * np.nanmean(relative_error_clip_c))\n", + " \n", + " if decay_c > best_decay:\n", + " best_decay = decay_c\n", + " best_code = candidate\n", + " \n", + " return best_code, best_decay" + ] + }, + { + "cell_type": "code", + "execution_count": 133, + "id": "ad49f064-7f5e-4b1a-bf5e-b8117182b1c1", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'200127201'" + ] + }, + "execution_count": 133, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "date_t = pd.Timestamp('2023-07-31')\n", + "date_t1 = pd.Timestamp('2023-06-30')\n", + "date_t0 = pd.Timestamp('2022-09-30')\n", + "\n", + "surgery_for_account('200127375',date_t,date_t1)[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 90, + "id": "f6b9d749-c0ae-444f-b373-1502a26b9c29", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
dateaccountscorescore_prevrelative_drop
558422015-01-312001274790.000000NaNNaN
554112015-02-282001274790.0000000.000000NaN
549802015-03-312001274790.0000000.000000NaN
545492015-04-302001274790.0000000.000000NaN
541182015-05-312001274790.0000000.000000NaN
..................
19672025-06-302001274790.0004860.0004854.028644e-04
15362025-07-312001274790.0004860.0004868.775873e-07
11052025-08-312001274790.0004860.0004863.857307e-05
6742025-09-302001274790.0004860.0004864.271008e-05
2432025-10-312001274790.0004860.0004861.099025e-03
\n", + "

130 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " date account score score_prev relative_drop\n", + "55842 2015-01-31 200127479 0.000000 NaN NaN\n", + "55411 2015-02-28 200127479 0.000000 0.000000 NaN\n", + "54980 2015-03-31 200127479 0.000000 0.000000 NaN\n", + "54549 2015-04-30 200127479 0.000000 0.000000 NaN\n", + "54118 2015-05-31 200127479 0.000000 0.000000 NaN\n", + "... ... ... ... ... ...\n", + "1967 2025-06-30 200127479 0.000486 0.000485 4.028644e-04\n", + "1536 2025-07-31 200127479 0.000486 0.000486 8.775873e-07\n", + "1105 2025-08-31 200127479 0.000486 0.000486 3.857307e-05\n", + "674 2025-09-30 200127479 0.000486 0.000486 4.271008e-05\n", + "243 2025-10-31 200127479 0.000486 0.000486 1.099025e-03\n", + "\n", + "[130 rows x 5 columns]" + ] + }, + "execution_count": 90, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "score_accounts_df[score_accounts_df['account']=='200127479']\n" + ] + }, + { + "cell_type": "code", + "execution_count": 127, + "id": "8b239ffa-2986-4212-af74-f0d4e1753748", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
dateaccountscorescore_prevrelative_drop
169572022-07-312001273750.0011440.0011400.003081
165262022-08-312001273750.0011440.0011440.000107
160952022-09-302001273750.0011440.0011440.000079
156642022-10-312001273750.0011500.0011440.004884
152332022-11-302001273750.0011500.0011500.000032
148022022-12-312001273750.0011500.0011500.000500
143712023-01-312001273750.0011590.0011500.007329
139402023-02-282001273750.0011610.0011590.001574
135092023-03-312001273750.0011610.0011610.000642
130782023-04-302001273750.0011620.0011610.000775
126472023-05-312001273750.0011640.0011620.001258
122162023-06-302001273750.0011640.0011640.000403
117852023-07-312001273750.0011770.0011640.011002
113542023-08-312001273750.0011770.0011770.000352
109232023-09-302001273750.0011770.0011770.000009
104922023-10-312001273750.0011790.0011770.000907
100612023-11-302001273750.0011800.0011790.000958
96302023-12-312001273750.0011810.0011800.001105
91992024-01-312001273750.0011880.0011810.005881
87682024-02-292001273750.0011900.0011880.001295
83372024-03-312001273750.0011900.0011900.000000
79062024-04-302001273750.0011940.0011900.003455
74752024-05-312001273750.0011940.0011940.000432
70442024-06-302001273750.0011940.0011940.000237
66132024-07-312001273750.0011970.0011940.001903
61822024-08-312001273750.0011970.0011970.000000
57512024-09-302001273750.0011970.0011970.000276
53202024-10-312001273750.0011990.0011970.001539
48892024-11-302001273750.0011990.0011990.000027
44582024-12-312001273750.0011990.0011990.000091
40272025-01-312001273750.0011990.0011990.000012
35962025-02-282001273750.0011990.0011990.000000
31652025-03-312001273750.0012000.0011990.000799
27342025-04-302001273750.0012000.0012000.000354
23032025-05-312001273750.0012000.0012000.000010
18722025-06-302001273750.0012010.0012000.000264
14412025-07-312001273750.0012020.0012010.000851
10102025-08-312001273750.0012020.0012020.000480
5792025-09-302001273750.0012020.0012020.000000
1482025-10-312001273750.0012030.0012020.000419
\n", + "
" + ], + "text/plain": [ + " date account score score_prev relative_drop\n", + "16957 2022-07-31 200127375 0.001144 0.001140 0.003081\n", + "16526 2022-08-31 200127375 0.001144 0.001144 0.000107\n", + "16095 2022-09-30 200127375 0.001144 0.001144 0.000079\n", + "15664 2022-10-31 200127375 0.001150 0.001144 0.004884\n", + "15233 2022-11-30 200127375 0.001150 0.001150 0.000032\n", + "14802 2022-12-31 200127375 0.001150 0.001150 0.000500\n", + "14371 2023-01-31 200127375 0.001159 0.001150 0.007329\n", + "13940 2023-02-28 200127375 0.001161 0.001159 0.001574\n", + "13509 2023-03-31 200127375 0.001161 0.001161 0.000642\n", + "13078 2023-04-30 200127375 0.001162 0.001161 0.000775\n", + "12647 2023-05-31 200127375 0.001164 0.001162 0.001258\n", + "12216 2023-06-30 200127375 0.001164 0.001164 0.000403\n", + "11785 2023-07-31 200127375 0.001177 0.001164 0.011002\n", + "11354 2023-08-31 200127375 0.001177 0.001177 0.000352\n", + "10923 2023-09-30 200127375 0.001177 0.001177 0.000009\n", + "10492 2023-10-31 200127375 0.001179 0.001177 0.000907\n", + "10061 2023-11-30 200127375 0.001180 0.001179 0.000958\n", + "9630 2023-12-31 200127375 0.001181 0.001180 0.001105\n", + "9199 2024-01-31 200127375 0.001188 0.001181 0.005881\n", + "8768 2024-02-29 200127375 0.001190 0.001188 0.001295\n", + "8337 2024-03-31 200127375 0.001190 0.001190 0.000000\n", + "7906 2024-04-30 200127375 0.001194 0.001190 0.003455\n", + "7475 2024-05-31 200127375 0.001194 0.001194 0.000432\n", + "7044 2024-06-30 200127375 0.001194 0.001194 0.000237\n", + "6613 2024-07-31 200127375 0.001197 0.001194 0.001903\n", + "6182 2024-08-31 200127375 0.001197 0.001197 0.000000\n", + "5751 2024-09-30 200127375 0.001197 0.001197 0.000276\n", + "5320 2024-10-31 200127375 0.001199 0.001197 0.001539\n", + "4889 2024-11-30 200127375 0.001199 0.001199 0.000027\n", + "4458 2024-12-31 200127375 0.001199 0.001199 0.000091\n", + "4027 2025-01-31 200127375 0.001199 0.001199 0.000012\n", + "3596 2025-02-28 200127375 0.001199 0.001199 0.000000\n", + "3165 2025-03-31 200127375 0.001200 0.001199 0.000799\n", + "2734 2025-04-30 200127375 0.001200 0.001200 0.000354\n", + "2303 2025-05-31 200127375 0.001200 0.001200 0.000010\n", + "1872 2025-06-30 200127375 0.001201 0.001200 0.000264\n", + "1441 2025-07-31 200127375 0.001202 0.001201 0.000851\n", + "1010 2025-08-31 200127375 0.001202 0.001202 0.000480\n", + "579 2025-09-30 200127375 0.001202 0.001202 0.000000\n", + "148 2025-10-31 200127375 0.001203 0.001202 0.000419" + ] + }, + "execution_count": 127, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_account.tail(40)" + ] + }, + { + "cell_type": "code", + "execution_count": 134, + "id": "92467723-1d7a-4eb7-bbf1-acb7ac60ce4a", + "metadata": {}, + "outputs": [], + "source": [ + "# Chirurgie des scores\n", + "\n", + "score_history_c = []\n", + "score_history_accounts_c = []\n", + "\n", + "scores = aum_account.set_index('Registrar Account - ID')['weight'].to_dict()\n", + "dic_account = {}\n", + "\n", + "score_history_c.append({\n", + " \"date\": ref_date,\n", + " \"total_score\": sum(scores.values())\n", + "})\n", + "\n", + "for account, score in scores.items():\n", + " score_history_accounts_c.append({\n", + " \"date\": ref_date,\n", + " \"account\": account,\n", + " \"score\": score\n", + " })\n", + "\n", + "dates_sorted = sorted(df['Centralisation Date'].unique(), reverse=True)\n", + "\n", + "alpha = 0.01\n", + "tol = 0.05\n", + "\n", + "for i in range(1, len(dates_sorted)):\n", + "\n", + " date_t = dates_sorted[i-1]\n", + " date_t1 = dates_sorted[i]\n", + "\n", + " new_scores = scores.copy()\n", + "\n", + " for account in scores.keys():\n", + " if account not in dic_account:\n", + " decay = score_decay_vector_max(account, date_t, date_t1, tol=tol, alpha=alpha)\n", + " if decay > 0.99: \n", + " new_scores[account] = scores[account] * decay\n", + " else:\n", + " chirurgie = surgery_for_account(account,date_t,date_t1, tol = 0.01, alpha = 0.01)\n", + " dic_account[account] = chirurgie[0]\n", + " new_scores[account] = scores[account] * chirurgie[1]\n", + " else: \n", + " decay = score_decay_vector_max(dic_account[account], date_t, date_t1, tol=tol, alpha=alpha)\n", + " if decay > 0.99: \n", + " new_scores[account] = scores[account] * decay\n", + " else:\n", + " chirurgie = surgery_for_account(dic_account[account],date_t,date_t1, tol = 0.01, alpha = 0.01)\n", + " dic_account[dic_account[account]] = chirurgie[0]\n", + " new_scores[account] = scores[account] * chirurgie[1]\n", + " scores = new_scores\n", + "\n", + " # score total\n", + " score_history_c.append({\n", + " \"date\": date_t1,\n", + " \"total_score\": sum(scores.values())\n", + " })\n", + "\n", + " # score par compte\n", + " for account, score in scores.items():\n", + " score_history_accounts_c.append({\n", + " \"date\": date_t1,\n", + " \"account\": account,\n", + " \"score\": score\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "68a2ff7e-a27b-4a6e-a22d-80787b88e17e", + "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": "3.13.12" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Essai.ipynb b/Essai.ipynb index 325e511..35e47a6 100644 --- a/Essai.ipynb +++ b/Essai.ipynb @@ -896,7 +896,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.11" + "version": "3.13.12" } }, "nbformat": 4, diff --git a/Migration.ipynb b/Migration.ipynb index bb6468f..fb85c11 100644 --- a/Migration.ipynb +++ b/Migration.ipynb @@ -1450,7 +1450,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.11" + "version": "3.13.12" } }, "nbformat": 4, diff --git a/rupture.ipynb b/rupture.ipynb index b3f74b9..b856349 100644 --- a/rupture.ipynb +++ b/rupture.ipynb @@ -1228,7 +1228,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.11" + "version": "3.13.12" } }, "nbformat": 4,