sthoumyre-ensae/Project_Carmignac
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

paco-dev #2

Merged
pgoze-ensae merged 39 commits from paco-dev into main 2026-04-13 21:51:04 +02:00
Conversation 0 Commits 39 Files Changed 25 +4589 -8
5 changed files with 4589 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit c6e8e97f49 - Show all commits

3
.gitignore vendored
View File

@ -1,3 +1,4 @@
data/ data/
data_exploration/ data_exploration/
notebooks/ notebooks/
*.csv

4529
carmignac_diagnostics.html Normal file
View File

File diff suppressed because one or more lines are too long

25
repair_challenge/push_s3.ipynb → push_s3.ipynb
View File

@ -2,7 +2,7 @@
"cells": [ "cells": [
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": 1,
"id": "5c8fc6c5", "id": "5c8fc6c5",
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
@ -24,19 +24,34 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": 2,
"id": "d43b725e", "id": "d43b725e",
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"push_file('AUM_repaired.csv', 'projet-bdc-carmignac-g3//paco/AUM_repaired.csv')\n", "push_file('carmignac_broken_months.csv', 'projet-bdc-carmignac-g3//paco/carmignac_broken_months.csv')\n",
"push_file('AUM_paths.csv', 'projet-bdc-carmignac-g3//paco/AUM_paths.csv')" "push_file('carmignac_error_account_agg.csv', 'projet-bdc-carmignac-g3//paco/carmignac_error_account_agg.csv')\n",
"push_file('carmignac_error_account.csv', 'projet-bdc-carmignac-g3//paco/carmignac_error_account.csv')"
] ]
} }
], ],
"metadata": { "metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": { "language_info": {
"name": "python" "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": 4,

4
repair_challenge/carmignac_repair.py
View File

@ -65,7 +65,7 @@ def load_broken_months(broken_months_path):
try: try:
df = pd.read_csv(broken_months_path, parse_dates=["date"]) df = pd.read_csv(broken_months_path, parse_dates=["date"])
broken = set(zip(pd.to_datetime(df["date"]), df["isin"].astype(str))) broken = set(zip(pd.to_datetime(df["date"]), df["isin"].astype(str)))
lags = set(zip(pd.to_datetime(df.loc[df["is_lag"], "date"]), lags = set(zip(pd.to_datetime(df.loc[df["is_lag"], "date"]),
df.loc[df["is_lag"], "isin"].astype(str))) df.loc[df["is_lag"], "isin"].astype(str)))
print(f"[Broken months] Loaded {len(broken)} flagged (isin, month) pairs " print(f"[Broken months] Loaded {len(broken)} flagged (isin, month) pairs "
f"({len(lags)} likely lags)") f"({len(lags)} likely lags)")
@ -801,7 +801,7 @@ def export_results(scores_history, mapping_history, surgery_log, all_months, out
# ───────────────────────────────────────────── # ─────────────────────────────────────────────
# 8. PIPELINE PRINCIPAL # 8. PIPELINE PRINCIPAL
# ───────────────────────────────────────────── # ─────────────────────────────────────────────
def run_pipeline(aum_path, flows_path, broken_months_path=None): def run_pipeline(aum_path, flows_path, broken_months_path="carmignac_broken_months.csv"):
print("=" * 60) print("=" * 60)
print("CARMIGNAC — Pipeline de réparation des Registrar IDs") print("CARMIGNAC — Pipeline de réparation des Registrar IDs")
print("=" * 60) print("=" * 60)

36
repair_challenge/diagnosis_logs.txt Normal file
View File

@ -0,0 +1,36 @@
============================================================
CARMIGNAC — Broken Months Diagnostics
============================================================
(isin, month) pairs examined : 39874
Broken (missing_pct > 5%) : 7545 (18.9%)
Of which likely lag : 491
Of which genuine gap : 7054
Top 10 by missing_pct:
date isin missing_flow missing_pct is_lag
2021-10-31 LU1299301876 400.000 2.000000 False
2019-07-31 LU0992625086 3940.000 1.995856 False
2021-02-28 LU0992626563 -681.559 1.995325 False
2019-01-31 FR0010148999 4595.753 1.994832 False
2020-08-31 LU0992631217 14928.331 1.991141 False
2020-11-30 LU1744630424 -35632.925 1.987345 False
2020-02-29 LU0992626480 25812.484 1.985526 False
2020-03-31 LU1299302098 930.648 1.982950 False
2020-07-31 LU1966631001 240.957 1.980076 False
2020-11-30 LU0807689822 -6181.993 1.972331 False
Most affected months:
n_broken total_missing
date
2019-12-31 123 2052282.985
2021-09-30 113 1892096.272
2020-12-31 107 652351.530
2020-11-30 105 833084.202
2021-01-31 104 784362.333
Aggregate broken months : 85 (of which lags: 2)
Max aggregate error stock : 3,057,936.2 shares (0.749% of total AUM)
[Export] Broken months CSV → carmignac_broken_months.csv
[Export] Error account (ISIN) → carmignac_error_account.csv
[Export] Error account (agg) → carmignac_error_account_agg.csv
[Export] HTML report → carmignac_diagnostics.html