Adriano/PythagorasGoal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(TP01): bug look-ahead ffill mixed-TF -> deploy a >=12h (1d), strategia DIFENSIVA

Browse Source 
Segnalato: ffill MIXED-TIMEFRAME su barre open-labeled (resample label="left") gonfiava il 4h
(~1.60 -> reale ~1.1). Ri-verifica per-SINGOLO-TF leak-free (guard prefix-recompute, leak=0 su
4h/6h/12h/1d): FULL Sh piatto ~1.3, hold-out 2025-26 MIGLIORE a 1d (Sh 0.31 / +3.5% vs buy&hold
-39%). Conclusione adottata: NON scendere sotto le 12h (sotto, costi+overfit dominano senza vantaggio).

- trend_portfolio.py: canonica PORT LF1d; resample_tf/resample_1d (resample_4h deprecato deploy);
  docstring con nota look-ahead + natura DIFENSIVA (taglia DD ~6x, non alpha).
- paper_trend.py: deploy a 1d (resample_1d, build_bars). 5 test passano.
- CLAUDE.md: TP01 ridescritta (>=12h/1d, gotcha ffill mixed-TF, difensiva).
- tp01_lowfreq.py + diario 2026-06-19-tp01-lookahead-fix-lf.md.
Gotcha: mai ffill/combine mixed-TF su timestamp open-labeled (close propagata indietro = leak).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Adriano Dal Pastro
2026-06-19 19:04:38 +00:00
parent 756a2bdf04
commit 12754c4908
5 changed files with 189 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/strategies/trend_portfolio.py
+24 -9
View File
@@ -4,13 +4,16 @@ Vincitrice della ricerca su dati certificati BTC/ETH (Deribit mainnet). TSMOM mu
(1-3-6 mesi) vol-targeted, portafoglio 50/50 BTC+ETH. Validata onestamente (no look-ahead,
fee 0.10% RT, positiva ogni anno 2019-2026, robusta su griglia e su tutti i timeframe 15m-1d).
Config canonica deployabile (PORT LF4h):
timeframe 4h, LONG-FLAT (niente short), vol-target 20%, leverage cap 2x.
-> CAGR ~16.6%, Sharpe ~1.32, maxDD ~12.3% (backtest 2019-2026 su 50/50 BTC+ETH).
Config canonica deployabile (PORT LF1d):
timeframe >=12h (1d RACCOMANDATO), LONG-FLAT (niente short), vol-target 20%, leverage cap 2x.
-> FULL Sharpe ~1.30, maxDD ~14%, HOLD-OUT 2025-26 Sharpe ~0.31 (calcolo per-TF leak-free).
Perche' long-flat e 4h: gli short del trend rendono meno e aggiungono DD; il 4h e' il punto
dolce (meno rumore/fee del 15m, meno lag dell'1d). Vedi docs/diary/2026-06-19-research-synthesis.md
e scripts/research/trackD_*.py.
NB LOOK-AHEAD (2026-06-19): un ffill MIXED-TIMEFRAME su barre open-labeled (label="left")
gonfiava il 4h (~1.60 -> reale ~1.1). Il calcolo per-SINGOLO-TF e' leak-free (guard
prefix-recompute), ma sotto le 12h costi+overfitting dominano SENZA vantaggio reale (FULL Sh
piatto ~1.3 da 12h a 4h; hold-out MIGLIORE a 1d). -> NON scendere sotto le 12h; deploy a 1d.
TP01 e' DIFENSIVA (taglia il DD ~6x vs buy&hold), NON alpha. Vedi
docs/diary/2026-06-19-tp01-lookahead-fix-lf.md e scripts/analysis/tp01_lowfreq.py.
API (tutto causale, decide con dati <= close[i]):
from src.strategies.trend_portfolio import TrendPortfolio, CANONICAL
@@ -168,16 +171,28 @@ def _bars_per_year(idx: pd.DatetimeIndex) -> float:
return 86400 * 365.25 / dt if dt and dt > 0 else 365.25
def resample_4h(df_1h: pd.DataFrame) -> pd.DataFrame:
"""Resample 1h -> 4h (confini 00:00 UTC). Schema con 'datetime'."""
def resample_tf(df_1h: pd.DataFrame, rule: str) -> pd.DataFrame:
"""Resample 1h -> rule (confini 00:00 UTC). Schema con 'datetime'.
NB: usare SOLO per-singolo-TF (qui leak-free); MAI ffill/combine mixed-TF su questi
timestamp open-labeled (label='left') -> look-ahead. Deploy a >=12h (vedi docstring modulo)."""
g = df_1h.copy()
idx = pd.to_datetime(g["timestamp"], unit="ms", utc=True)
idx.name = "dt"
g.index = idx
out = g.resample("4h", label="left", closed="left").agg(
out = g.resample(rule, label="left", closed="left").agg(
{"open": "first", "high": "max", "low": "min", "close": "last", "volume": "sum"})
out = out.dropna(subset=["open"])
out["datetime"] = out.index
epoch = pd.Timestamp("1970-01-01", tz="UTC")
out["timestamp"] = ((out.index - epoch) // pd.Timedelta(milliseconds=1)).astype("int64")
return out.reset_index(drop=True)[["timestamp", "open", "high", "low", "close", "volume", "datetime"]]
def resample_1d(df_1h: pd.DataFrame) -> pd.DataFrame:
"""TF canonico di deploy (>=12h). Resample 1h -> 1d."""
return resample_tf(df_1h, "1D")
def resample_4h(df_1h: pd.DataFrame) -> pd.DataFrame:
"""DEPRECATO per il deploy (sotto le 12h: costi+overfit dominano). Retro-compat ricerca."""
return resample_tf(df_1h, "4h")