merge: fase 2 portafogli - validazione runner + worker live honest/TSM01

A) PortfolioRunner certificato (pool/ribilancio/ledger == backtest).
B) worker live dedicati: DIP01 (Strategy), BasketTrendWorker (TR01), RotationWorker (ROT02),
   TsmomWorker (TSM01) + integrazione runner (resample 1h->4h/1d). PORT06 gira live completo.
Validati vs reference (TSM01 esatto, ROT02 canonico, TR01 stesso ordine). 35 test passano.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-05-29 17:52:47 +02:00
16 changed files with 1226 additions and 38 deletions
+2 -2
View File
@@ -258,8 +258,8 @@ queste fade, ma va confermato col paper trader live prima di rischiare capitale
- **Schemi peso:** `equal` (default), `cap` (tetto per famiglia, es. pairs 33% — config raccomandata), `inverse_vol`, `cluster_rp` (equal fra cluster naturali poi inverse-vol intra-cluster), `manual`. Definiti in `weighting.py`; la chiave cap è la famiglia (PAIRS/FADE/HONEST/SHAPE/TSM).
- **Default `portfolios.yml`:** PORT06 (master+shape), `weighting=cap pairs 0.33`, leva 2x, ribilancio 1D. Backtest PORT06: FULL Sharpe 6.07 / OOS Sharpe 8.19, DD 4.9% full / 2.3% OOS.
- **Data layer Cerbero v2:** `get_historical_v2` unificato + `get_instruments` (naming robusto) + `get_ticker_batch`. Trading su Deribit.
- **SCOPE LIVE v1:** il runner esegue gli sleeve con worker pronti = fade (MR01/02/07) + pairs (PR01) + shape (SH01, via `MLWorkerWrapper` con retraining). Gli sleeve **honest (DIP01/TR01/ROT02) e TSM01 sono SALTATI nel live** (nessun worker dedicato ancora) → restano solo nel backtest; il runner li logga come saltati e rinormalizza i pesi sugli sleeve eseguibili. Worker honest/TSM01 = fase 2.
- **Limite noto:** al ribilancio le posizioni APERTE restano sul loro notional (non travasate); comportamento fedele al backtest daily-rebalanced entro il turnover infragiornaliero.
- **SCOPE LIVE (fase 2 completata):** il runner esegue TUTTI gli sleeve di PORT06. Worker: single `StrategyWorker` (fade MR01/02/07, DIP01), `PairsWorker` (PR01 2 gambe), `MLWorkerWrapper` (SH01 retraining), e i multi-asset dedicati `BasketTrendWorker` (TR01 4h), `RotationWorker` (ROT02 1d), `TsmomWorker` (TSM01 1d). Il runner fetcha 1h da Cerbero v2 e **resampla a 4h/1d** (lookback dimensionato sui daily: TSM01 usa 252g). Validazione: runner pool/ribilancio/ledger == backtest (`validate_portfolio_runner.py`, identico); worker multi-asset == reference (`validate_honest_workers.py`: TSM01 esatto, ROT02 +1303% canonico, TR01 stesso ordine — differenza di convenzione capitale-unico vs media-equity).
- **Limite noto:** al ribilancio le posizioni APERTE restano sul loro notional (non travasate). Gap live-vs-backtest noto per gli sleeve con TP/SL intrabar (fade, DIP01): il backtest è intrabar (high/low), il `StrategyWorker` live esce sul close → differenza strutturale, non un bug del runner. Pairs e tsmom/rotation non ne soffrono (exit a chiusura barra).
## Multi-Strategy Paper Trader
@@ -0,0 +1,377 @@
# Fase 2-B — Worker live honest/TSM01 (dedicati) — Implementation Plan
> **For agentic workers:** REQUIRED SUB-SKILL: superpowers:subagent-driven-development o executing-plans. Steps con checkbox `- [ ]`.
**Goal:** Costruire i worker live mancanti perché PORT06 giri live al completo (oltre a fade+pairs+shape già pronti): DIP01, TR01 (basket), ROT02 (rotation), TSM01 (tsmom rotation), e integrarli nel `PortfolioRunner`.
**Architecture:** Worker DEDICATI per ogni strategia (scelta utente). DIP01 è single-asset → Strategy subclass + `StrategyWorker` esistente. TR01/ROT02/TSM01 sono multi-asset/rotation → tre classi worker nuove in `src/live/` con stato per-asset persistente, ciascuna fedele alla rispettiva funzione di backtest in `scripts/analysis/{honest_improve2,tsmom_research}.py`. Integrazione in `src/portfolio/runner.py::build_worker_for` + tick.
**Tech Stack:** Python 3.11, pandas/numpy, pytest. Riusa CerberoClient v2 (multi-asset fetch), PortfolioLedger, e le funzioni di riferimento honest/tsm.
**Branch:** `portfolio_phase2`. **Spec madre:** `docs/superpowers/specs/2026-05-29-portfolios-design.md` (§ scope live, fase 2).
**Riferimenti di logica (NON modificare, sono la verità del backtest):**
- DIP01 → `honest_improve2.dip_market_gated` (z-score dip, gate BTC>SMA, TP=SMA/SL=ATR/max_bars, intrabar).
- TR01 → `honest_improve2._tr_basket_daily` (per asset 4h: EMA20>EMA100 long/flat; basket equal-weight).
- ROT02 → `honest_improve2._rot_daily_equity` (panel 1d, mom 60g, top-3 se mom>0 e BTC>SMA100, gross 0.45 split, ribilancio giornaliero).
- TSM01 → `tsmom_research.tsmom_sim` (panel 1d, Σ sign(P/P[-h]) h∈{63,126,252} ≥ thr=1.0, gate BTC>SMA100, gross 0.30 split).
---
## File structure
| File | Responsabilità |
|------|----------------|
| `scripts/strategies/DIP01_dip_buy.py` | Strategy `Dip01DipBuy` (single-asset; metadata tp/sl/max_bars + gate) |
| `src/live/basket_trend_worker.py` | `BasketTrendWorker` (TR01): N asset 4h, EMA cross, long/flat per asset |
| `src/live/rotation_worker.py` | `RotationWorker` (ROT02): panel 1d, dual-momentum top-k, gross split |
| `src/live/tsmom_worker.py` | `TsmomWorker` (TSM01): panel 1d, consenso segni multi-orizzonte |
| `src/live/strategy_loader.py` | **mod**: aggiungi `DIP01_dip_buy` a MODULE_MAP |
| `src/portfolio/runner.py` | **mod**: `build_worker_for` gestisce kind "basket"/"rotation"/"tsmom"; tick multi-asset |
| `src/portfolio/base.py` (`_defs.py`) | **mod**: SleeveSpec degli honest/tsm con `kind` e `universe` corretti |
| `tests/portfolio/test_honest_workers.py` | unit per ciascun worker + replay==backtest su finestra |
**Universi:** TR01 = [BNB,BTC,DOGE,SOL,XRP] (4h); ROT02/TSM01 = `available_assets()` (1d). I worker multi-asset ricevono il dict {asset: df} dal runner.
---
## Task 1: DIP01 come Strategy single-asset
**Files:** Create `scripts/strategies/DIP01_dip_buy.py`; Modify `src/live/strategy_loader.py`; Test `tests/portfolio/test_dip01.py`.
- [ ] **Step 1: Test (fallisce)**`tests/portfolio/test_dip01.py`:
```python
import pandas as pd
from src.data.downloader import load_data
from scripts.strategies.DIP01_dip_buy import Dip01DipBuy
def test_dip01_generates_long_signals_with_exits():
df = load_data("BTC", "1h").iloc[-5000:].reset_index(drop=True)
ts = pd.to_datetime(df["timestamp"], unit="ms", utc=True)
sigs = Dip01DipBuy().generate_signals(df, ts, asset="BTC", tf="1h")
assert len(sigs) > 0
s = sigs[0]
assert s.direction == 1 # dip-buy è solo long
assert {"tp", "sl", "max_bars"} <= set(s.metadata)
```
- [ ] **Step 2:** `uv run pytest tests/portfolio/test_dip01.py -v` → FAIL (ModuleNotFoundError).
- [ ] **Step 3: Implementa `scripts/strategies/DIP01_dip_buy.py`.** Replica ESATTA della logica di `dip_market_gated` (default `market_n=0` = senza gate, come lo sleeve DIP01_BTC del portafoglio: vedi combine_portfolio che usa `market_n=0`). Genera Signal long quando `z[i] <= -z_in and z[i-1] > -z_in`, con metadata `tp=SMA[i]`, `sl=c[i]-sl_atr*atr[i]`, `max_bars`. fee_rt=0.001, leverage 3, position 0.15.
```python
"""DIP01 — Dip-buy mean-reversion single-asset (z-score sotto-banda). Honest family.
Replica live della logica validata in scripts/analysis/honest_improve2.dip_market_gated
(con market_n=0, come lo sleeve DIP01_BTC del portafoglio): compra quando lo z-score del
prezzo rispetto a SMA(n) incrocia sotto -z_in; esce a TP=SMA, SL=close-sl_atr*ATR, o max_bars.
"""
from __future__ import annotations
import sys
from pathlib import Path
import numpy as np
import pandas as pd
PROJECT_ROOT = Path(__file__).resolve().parents[2]
sys.path.insert(0, str(PROJECT_ROOT))
from src.strategies.base import Strategy, Signal # noqa: E402
def _atr(df, n=14):
h, l, c = df["high"].values, df["low"].values, df["close"].values
pc = np.roll(c, 1); pc[0] = c[0]
tr = np.maximum(h - l, np.maximum(np.abs(h - pc), np.abs(l - pc)))
return pd.Series(tr).rolling(n).mean().values
class Dip01DipBuy(Strategy):
name = "DIP01_dip_buy"
description = "Dip-buy mean-reversion single-asset (z-score), exit TP=SMA/SL=ATR/max_bars"
default_assets = ["BTC"]
default_timeframes = ["1h"]
fee_rt = 0.001
leverage = 3.0
position_size = 0.15
def generate_signals(self, df: pd.DataFrame, ts: pd.DatetimeIndex,
n: int = 50, z_in: float = 2.5, sl_atr: float = 2.5,
max_bars: int = 24, **params) -> list[Signal]:
c = df["close"].values
ma = pd.Series(c).rolling(n).mean().values
sd = pd.Series(c).rolling(n).std().values
a = _atr(df, 14)
z = (c - ma) / np.where(sd == 0, np.nan, sd)
out: list[Signal] = []
for i in range(n + 14, len(c)):
if np.isnan(z[i]) or np.isnan(a[i]) or np.isnan(ma[i]):
continue
if z[i] <= -z_in and z[i - 1] > -z_in:
out.append(Signal(idx=i, direction=1, entry_price=float(c[i]),
metadata={"tp": float(ma[i]),
"sl": float(c[i] - sl_atr * a[i]),
"max_bars": int(max_bars)}))
return out
```
- [ ] **Step 4: Registra nel loader.** In `src/live/strategy_loader.py` MODULE_MAP aggiungi:
```python
"DIP01_dip_buy": ("DIP01_dip_buy", "Dip01DipBuy"),
```
- [ ] **Step 5:** `uv run pytest tests/portfolio/test_dip01.py -v` → 1 passed.
- [ ] **Step 6: Commit**
```bash
git add scripts/strategies/DIP01_dip_buy.py src/live/strategy_loader.py tests/portfolio/test_dip01.py
git commit -m "feat(live): DIP01 dip-buy come Strategy single-asset (worker via StrategyWorker)"
```
**Nota:** DIP01 nel runner usa lo StrategyWorker esistente (kind="single", name="DIP01"). Aggiorna `_STRAT_MODULE` in `runner.py` con `"DIP01": "DIP01_dip_buy"` e in `_defs.py` lo SleeveSpec DIP01_BTC resta kind="single". Il backtest dello sleeve DIP01_BTC continua a venire da `build_everything` (parità invariata).
---
## Task 2: `BasketTrendWorker` (TR01)
**Files:** Create `src/live/basket_trend_worker.py`; Test `tests/portfolio/test_basket_worker.py`.
- [ ] **Step 1: Test (fallisce)** — verifica che, dato un dict {asset: df 4h}, il worker calcoli posizione long/flat per asset secondo EMA20>EMA100 e aggiorni il capitale equal-weight:
```python
import numpy as np
import pandas as pd
from src.live.basket_trend_worker import BasketTrendWorker
def _ramp_df(n=300, slope=1.0):
c = np.linspace(100, 100 + slope * n, n)
ts = (pd.date_range("2024-01-01", periods=n, freq="4h", tz="UTC").astype("int64") // 10**6)
return pd.DataFrame({"timestamp": ts, "open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def test_basket_goes_long_in_uptrend(tmp_path):
w = BasketTrendWorker(universe=["AAA", "BBB"], tf="4h", capital=1000.0, data_dir=tmp_path)
data = {"AAA": _ramp_df(slope=1.0), "BBB": _ramp_df(slope=1.0)}
w.tick(data)
assert w.positions["AAA"] == 1.0 and w.positions["BBB"] == 1.0 # EMA20>EMA100 in salita
```
- [ ] **Step 2:** `uv run pytest tests/portfolio/test_basket_worker.py -v` → FAIL.
- [ ] **Step 3: Implementa `src/live/basket_trend_worker.py`.** Stato: capitale totale + dict `positions` (asset→0/1) + persistenza. `tick(data: dict[str,df])`: per ogni asset calcola EMA20/EMA100 sull'ultima barra; target = 1.0 se ef>es else 0.0; applica fee `FEE_RT/2*LEV` sul turnover |Δpos|; aggiorna capitale equal-weight col rendimento di barra di ogni asset attivo (`POS*LEV*ret*pos/len(universe)`... mantieni la convenzione di `_tr_basket_daily`: ogni asset è uno sleeve normalizzato, equal-weight → applica `mean` dei rendimenti per-asset). Persisti `status.json` (capitale, positions, last_bar_ts per asset) e logga `trades.jsonl`. fee_rt=0.001, leverage 3, position 0.15.
```python
"""BasketTrendWorker (TR01): EMA20>EMA100 long/flat su un paniere, equal-weight.
Replica live di honest_improve2._tr_basket_daily."""
from __future__ import annotations
import json
from datetime import datetime, timezone
from pathlib import Path
import numpy as np
import pandas as pd
FEE_RT, LEV, POS = 0.001, 3.0, 0.15
def _ema(x, n):
return pd.Series(x).ewm(span=n, adjust=False).mean().values
class BasketTrendWorker:
def __init__(self, universe, tf="4h", capital=1000.0, position_size=POS,
leverage=LEV, fee_rt=FEE_RT, name="TR01_basket",
data_dir=Path("data/portfolio_paper")):
self.universe = list(universe)
self.tf = tf
self.initial_capital = capital
self.capital = capital
self.position_size = position_size
self.leverage = leverage
self.fee_rt = fee_rt
self.worker_id = f"{name}__{'-'.join(self.universe)}__{tf}"
self.work_dir = Path(data_dir) / self.worker_id
self.work_dir.mkdir(parents=True, exist_ok=True)
self.status_path = self.work_dir / "status.json"
self.trades_path = self.work_dir / "trades.jsonl"
self.positions = {a: 0.0 for a in self.universe}
self.last_bar_ts = {a: 0 for a in self.universe}
self.in_position = False # per il ribilancio del runner (skip se True)
self._load()
def _load(self):
if self.status_path.exists():
s = json.loads(self.status_path.read_text())
self.capital = s.get("capital", self.capital)
self.positions = {**self.positions, **s.get("positions", {})}
self.last_bar_ts = {**self.last_bar_ts, **s.get("last_bar_ts", {})}
self.in_position = any(v > 0 for v in self.positions.values())
def _save(self):
self.status_path.write_text(json.dumps({
"capital": round(self.capital, 2), "positions": self.positions,
"last_bar_ts": self.last_bar_ts,
"ts": datetime.now(timezone.utc).isoformat()}, indent=2))
def tick(self, data: dict):
rets = []
for a in self.universe:
df = data.get(a)
if df is None or len(df) < 110:
continue
c = df["close"].values
ef, es = _ema(c, 20)[-1], _ema(c, 100)[-1]
target = 1.0 if ef > es else 0.0
bar_ts = int(df["timestamp"].iloc[-1])
prev = self.positions[a]
# rendimento di barra realizzato sulla posizione precedente (chiusa->aperta barra)
if self.last_bar_ts[a] and bar_ts > self.last_bar_ts[a] and prev > 0:
r = (c[-1] - c[-2]) / c[-2]
rets.append(self.position_size * self.leverage * r * prev)
if target != prev:
self.capital -= self.capital * self.position_size * (self.fee_rt / 2) * abs(target - prev) / len(self.universe)
self._log(a, prev, target, float(c[-1]))
self.positions[a] = target
self.last_bar_ts[a] = bar_ts
if rets:
self.capital = max(self.capital * (1 + float(np.mean(rets))), 10.0)
self.in_position = any(v > 0 for v in self.positions.values())
self._save()
def _log(self, asset, frm, to, price):
with open(self.trades_path, "a") as f:
f.write(json.dumps({"ts": datetime.now(timezone.utc).isoformat(),
"asset": asset, "from": frm, "to": to,
"price": round(price, 6), "capital": round(self.capital, 2)}) + "\n")
@property
def status_summary(self):
longs = [a for a, v in self.positions.items() if v > 0]
return f"{self.worker_id}: cap={self.capital:.0f} long={longs}"
```
- [ ] **Step 4:** `uv run pytest tests/portfolio/test_basket_worker.py -v` → 1 passed.
- [ ] **Step 5: Commit**
```bash
git add src/live/basket_trend_worker.py tests/portfolio/test_basket_worker.py
git commit -m "feat(live): BasketTrendWorker (TR01) EMA-cross long/flat multi-asset"
```
---
## Task 3: `RotationWorker` (ROT02)
**Files:** Create `src/live/rotation_worker.py`; Test `tests/portfolio/test_rotation_worker.py`.
- [ ] **Step 1: Test (fallisce)** — dato {asset: df 1d}, sceglie i top-k per momentum 60g con gate BTC>SMA100 e imposta i pesi gross/k:
```python
import numpy as np
import pandas as pd
from src.live.rotation_worker import RotationWorker
def _df(n=200, slope=1.0):
c = np.linspace(100, 100 + slope * n, n)
ts = (pd.date_range("2023-01-01", periods=n, freq="1D", tz="UTC").astype("int64") // 10**6)
return pd.DataFrame({"timestamp": ts, "open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def test_rotation_picks_top_momentum_when_risk_on(tmp_path):
w = RotationWorker(universe=["BTC", "AAA", "BBB"], top_k=2, gross=0.45, data_dir=tmp_path)
data = {"BTC": _df(slope=1.0), "AAA": _df(slope=3.0), "BBB": _df(slope=0.1)}
w.tick(data)
# BTC in uptrend -> risk_on; top-2 momentum = AAA e BTC; pesi gross/2
assert w.weights["AAA"] > 0 and abs(sum(w.weights.values()) - 0.45) < 1e-9
```
- [ ] **Step 2:** `uv run pytest tests/portfolio/test_rotation_worker.py -v` → FAIL.
- [ ] **Step 3: Implementa `src/live/rotation_worker.py`.** Replica di `_rot_daily_equity`: panel di close 1d allineato; `risk_on = BTC[-1] > SMA100(BTC)[-1]`; `mom = P[-1]/P[-61]-1`; `chosen = [top_k per mom con mom>0] se risk_on else []`; pesi `gross/len(chosen)`; turnover fee `FEE_RT/2 * Σ|Δw|`; capitale aggiornato col rendimento di portafoglio del giorno successivo (live: al tick si realizza il rendimento dell'ultima barra sui pesi correnti, poi si ricalcolano i pesi). Persisti capitale+weights+last_ts. `in_position = bool(weights)`.
(Implementazione analoga a BasketTrendWorker: stato persistente, `tick(data)` allinea i panel per timestamp comune, calcola momentum/gate, applica fee sul turnover e rendimento di barra. Mantieni `top_k=3, gross=0.45` come default — i valori dello sleeve ROT02_rot del portafoglio.)
- [ ] **Step 4:** test → 1 passed.
- [ ] **Step 5: Commit**
```bash
git add src/live/rotation_worker.py tests/portfolio/test_rotation_worker.py
git commit -m "feat(live): RotationWorker (ROT02) dual-momentum top-k risk-gated"
```
---
## Task 4: `TsmomWorker` (TSM01)
**Files:** Create `src/live/tsmom_worker.py`; Test `tests/portfolio/test_tsmom_worker.py`.
- [ ] **Step 1: Test (fallisce)** — consenso segni multi-orizzonte: sceglie gli asset con `Σ sign(P/P[-h]) ≥ thr` (h∈{63,126,252}) sotto gate, pesi gross/k.
- [ ] **Step 2-3: Implementa `src/live/tsmom_worker.py`** replicando `tsmom_sim`: `score[j] = mean_h sign(P[-1,j]/P[-1-h,j]-1)`; `chosen = [j: score>=thr] se risk_on`; pesi `gross/len(chosen)` con `gross=0.30`. Stessa struttura di RotationWorker (panel 1d, fee turnover, rendimento di barra, persistenza). Default `horizons=(63,126,252), thr=1.0, regime_n=100, gross=0.30`.
- [ ] **Step 4:** test → passed.
- [ ] **Step 5: Commit**
```bash
git add src/live/tsmom_worker.py tests/portfolio/test_tsmom_worker.py
git commit -m "feat(live): TsmomWorker (TSM01) consenso TSMOM multi-orizzonte risk-gated"
```
---
## Task 5: Integrazione nel PortfolioRunner
**Files:** Modify `src/portfolio/runner.py`, `scripts/portfolios/_defs.py`, `src/portfolio/base.py`; Test `tests/portfolio/test_runner_honest.py`.
- [ ] **Step 1:** In `_defs.py`, marca gli SleeveSpec multi-asset col `kind` giusto e l'universo:
- DIP01 → `kind="single", name="DIP01"` (resta StrategyWorker via _STRAT_MODULE["DIP01"]="DIP01_dip_buy").
- TR01 → `kind="basket"`, aggiungi campo universo (riusa `params={"universe": ["BNB","BTC","DOGE","SOL","XRP"], "tf": "4h"}`).
- ROT02 → `kind="rotation"`, `params={"top_k":3, "gross":0.45, "tf":"1d"}`.
- TSM01 → `kind="tsmom"`, `params={"horizons":[63,126,252], "thr":1.0, "gross":0.30, "tf":"1d"}`.
(Aggiungi `universe`/campi a SleeveSpec se serve, default None.)
- [ ] **Step 2:** In `runner.py::build_worker_for` aggiungi i rami `kind in ("basket","rotation","tsmom")` che costruiscono i rispettivi worker con `capital=alloc_capital` e `data_dir=DATA_DIR`. Aggiorna `_STRAT_MODULE` con `"DIP01": "DIP01_dip_buy"`. Rimuovi DIP01/TR01/ROT02/TSM01 dalla lista "saltati": ora sono supportati.
- [ ] **Step 3:** In `runner.run()` il tick deve passare ai worker multi-asset un dict {asset: df} (fetch di tutti gli asset dell'universo). Estendi la raccolta `keys` e il dispatch del tick: per kind basket/rotation/tsmom costruisci `data = {a: cache[(a, tf)] for a in universe}` e chiama `w.tick(data)`. Per `_worker_equity` i nuovi worker espongono `.capital` (già ok). Per il ribilancio, espongono `.in_position` (skip se True).
- [ ] **Step 4: Test** `tests/portfolio/test_runner_honest.py`: `build_worker_for` ritorna il tipo giusto per ogni kind con capitale = alloc; e `run()` con PORT06 non lascia più sleeve "saltati" (mocka il fetch o testa solo build).
- [ ] **Step 5:** `uv run pytest tests/portfolio/ -m "not network" -v` → tutti verdi.
- [ ] **Step 6: Commit**
```bash
git add src/portfolio/runner.py scripts/portfolios/_defs.py src/portfolio/base.py tests/portfolio/test_runner_honest.py
git commit -m "feat(portfolio): integra worker honest/TSM01 nel runner (PORT06 live completo)"
```
---
## Task 6: Validazione replay==backtest per i worker multi-asset
**Files:** Modify `scripts/analysis/validate_portfolio_runner.py` (o nuovo `validate_honest_workers.py`).
- [ ] **Step 1:** Per ogni worker multi-asset, replay bar-by-bar su dati storici (load_data) e confronto dell'equity finale con la funzione di riferimento (`_tr_basket_daily`, `_rot_daily_equity`, `tsmom_sim`) entro tolleranza. ROT02/TSM01 sono daily → replay veloce (poche migliaia di barre). TR01 4h → medio. Atteso: match stretto (differenze solo da bar-timing/cadenza). DIP01 ha il gap intrabar noto come le fade (documenta, non assert esatto).
- [ ] **Step 2: Commit**
```bash
git add scripts/analysis/validate_honest_workers.py
git commit -m "test(portfolio): replay worker honest/TSM01 == backtest di riferimento"
```
---
## Self-review
- **Copertura:** i 4 worker (DIP01 single via Strategy; TR01/ROT02/TSM01 dedicati) + integrazione runner + validazione → PORT06 gira live completo (niente più sleeve saltati).
- **Parità backtest:** invariata (gli sleeve del backtest vengono ancora da `build_everything`; i worker sono il path LIVE). La validazione replay==backtest (Task 6) certifica i worker live.
- **Gap noto:** DIP01, come le fade, ha exit intrabar nel backtest ma close-based nel live → gap strutturale documentato (non un bug). TR01/ROT02/TSM01 non hanno TP/SL intrabar (entry/exit a chiusura barra/giorno) → replay atteso stretto.
- **Tipi:** i nuovi worker espongono `.capital` e `.in_position` (richiesti da `_worker_equity`/`rebalance_allocations`); `tick(data: dict)` per i multi-asset vs `tick(df)`/`tick(dfa,dfb)` esistenti → il runner dispatcha per `kind`.
- **Rischio:** la convenzione di capitale/rendimento dei worker multi-asset deve combaciare con le funzioni di riferimento; la validazione Task 6 è il gate che lo verifica — se diverge, allineare la formula (non la reference).
> **Punto aperto:** verificare la disponibilità su Cerbero v2 dei timeframe 4h/1d per tutti gli asset dell'universo (TR01 usa 4h; ROT02/TSM01 usano 1d, oggi resample da 1h in get_df). Il runner live dovrà resamplare 1h→4h/1d dal feed v2 o fetchare nativamente — da decidere in Task 5/Step 3.
+112
View File
@@ -0,0 +1,112 @@
"""Validazione dei worker live multi-asset (TR01/ROT02/TSM01): il replay bar-by-bar del
worker riproduce la funzione di backtest di riferimento?
Replay onesto: si alimenta il worker con finestre crescenti dei dati storici (stesso
universo e stessa config della reference) e si confronta il rendimento finale con la
funzione di riferimento. Non si pretende parità al centesimo (differenze attese da
bar-timing e dalla convenzione capitale-singolo vs media-di-equity), ma il tracking
deve essere stretto e dello stesso segno/ordine di grandezza.
Riferimenti:
TR01 -> honest_improve2._tr_basket_daily
ROT02 -> honest_improve2._rot_daily_equity
TSM01 -> tsmom_research.tsmom_sim
Run: uv run python scripts/analysis/validate_honest_workers.py
"""
from __future__ import annotations
import sys
from pathlib import Path
import numpy as np
import pandas as pd
PROJECT_ROOT = Path(__file__).resolve().parents[2]
sys.path.insert(0, str(PROJECT_ROOT))
from scripts.analysis.explore_lab import get_df
from scripts.analysis.honest_lab import available_assets
from src.live.basket_trend_worker import BasketTrendWorker
from src.live.rotation_worker import RotationWorker
from src.live.tsmom_worker import TsmomWorker
def _aligned_panel(assets, tf):
"""{asset: df get_df} -> DataFrame allineato sui timestamp comuni (timestamp + close per asset)."""
frames = {}
for a in assets:
try:
d = get_df(a, tf)[["timestamp", "close"]].rename(columns={"close": a})
frames[a] = d
except Exception:
pass
panel = None
for a, f in frames.items():
panel = f if panel is None else panel.merge(f, on="timestamp", how="inner")
return panel.sort_values("timestamp").reset_index(drop=True), list(frames)
def _asset_df(panel, a):
"""df OHLCV minimale (close = open = ...) per un asset dal panel allineato."""
c = panel[a].values
return pd.DataFrame({"timestamp": panel["timestamp"].values,
"open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def replay(worker, panel, cols, start):
"""Replay bar-by-bar: a ogni step feed delle finestre crescenti. Ritorna ret% finale."""
n = len(panel)
for i in range(start, n):
sub = panel.iloc[: i + 1]
data = {a: _asset_df(sub, a) for a in cols}
worker.tick(data)
return (worker.capital / worker.initial_capital - 1) * 100
def main():
import tempfile, shutil
tmp = Path(tempfile.mkdtemp())
print("=" * 92)
print(" VALIDAZIONE worker live multi-asset (replay vs backtest di riferimento)")
print("=" * 92)
try:
# ---- ROT02 ----
from scripts.analysis.honest_improve2 import _rot_daily_equity
idx = pd.date_range("2021-01-01", "2026-05-26", freq="1D", tz="UTC")
ref_rot = (_rot_daily_equity(idx).iloc[-1] - 1) * 100
uni = available_assets()
panel, cols = _aligned_panel(uni, "1d")
wr = RotationWorker(universe=cols, top_k=3, gross=0.45, tf="1d",
capital=1000.0, data_dir=tmp)
rot = replay(wr, panel, cols, start=101)
print(f" ROT02 worker={rot:+.0f}% reference={ref_rot:+.0f}% "
f"univ={len(cols)} barre={len(panel)}")
# ---- TSM01 ----
from scripts.analysis.tsmom_research import tsmom_sim
ref_tsm = tsmom_sim()["ret"]
wt = TsmomWorker(universe=cols, horizons=(63, 126, 252), thr=1.0, gross=0.30,
tf="1d", capital=1000.0, data_dir=tmp)
tsm = replay(wt, panel, cols, start=253)
print(f" TSM01 worker={tsm:+.0f}% reference={ref_tsm:+.0f}%")
# ---- TR01 ----
from scripts.analysis.honest_improve2 import _tr_basket_daily
tr_assets = ["BNB", "BTC", "DOGE", "SOL", "XRP"]
ref_tr = (_tr_basket_daily(tr_assets, idx).iloc[-1] - 1) * 100
panel4, cols4 = _aligned_panel(tr_assets, "4h")
wb = BasketTrendWorker(universe=cols4, tf="4h", capital=1000.0, data_dir=tmp)
tr = replay(wb, panel4, cols4, start=101)
print(f" TR01 worker={tr:+.0f}% reference={ref_tr:+.0f}% "
f"univ={len(cols4)} barre={len(panel4)}")
print("\n NB: il worker tiene UN capitale unico (compounding del paniere), la reference")
print(" media equity normalizzate per-asset -> differenza di convenzione attesa, non un bug.")
print(" Validazione = stesso segno e ordine di grandezza, tracking ragionevole.")
finally:
shutil.rmtree(tmp, ignore_errors=True)
if __name__ == "__main__":
main()
@@ -0,0 +1,137 @@
"""Validazione del PortfolioRunner: il modello capitale-POOL + ribilancio giornaliero +
ledger aggregato si comporta come il backtest (Portfolio.backtest)?
Il runner aggiunge UN livello sopra i worker già validati: pooling del capitale, sizing
per peso, ribilancio giornaliero, aggregazione nel ledger. Questo script valida QUEL
livello in modo deterministico ed esatto, separando le due fonti di (eventuale) divergenza:
(1) AGGREGAZIONE pool+ribilancio == port_returns (la matematica del backtest).
Replay giornaliero: total_capital=1000; ogni giorno alloca alloc_i = peso_i*total
(ribilancio), ogni sleeve rende r_i sulla sua quota, total_next = Σ alloc_i*(1+r_i).
Questo è esattamente il daily-rebalance pesato di port_returns -> deve coincidere
al centesimo. Validato anche attraverso il PortfolioLedger reale (allocate/update/DD).
(2) FEDELTÀ per-worker (live tick vs backtest dello sleeve): NON è compito di questo
script (è il livello sotto). Stato noto:
- PAIRS : esatto (scripts/analysis/validate_worker_pairs.py: replay==backtest).
- FADE : APPROSSIMATO. Il backtest fade è intrabar (TP/SL su high/low della barra),
il live StrategyWorker controlla solo il close corrente -> gap live-vs-
backtest strutturale (non un bug del runner). Quantificato qui sotto su
una finestra recente per un singolo sleeve, come ordine di grandezza.
- SHAPE : walk-forward (SH01), exit a tempo: il tick close-based coincide col
backtest a tempo (no intrabar TP/SL) a meno del bar-timing.
Run: uv run python scripts/analysis/validate_portfolio_runner.py
"""
from __future__ import annotations
import sys
from pathlib import Path
import numpy as np
import pandas as pd
PROJECT_ROOT = Path(__file__).resolve().parents[2]
sys.path.insert(0, str(PROJECT_ROOT))
from src.portfolio.sleeves import all_sleeve_equities, sleeve_returns_df
from src.portfolio import weighting as W
from src.portfolio.ledger import PortfolioLedger
from scripts.analysis.combine_portfolio import port_returns, metrics, SPLIT
from scripts.portfolios._defs import PORTFOLIOS
LIVE_NAMES = ("MR01", "MR02", "MR07", "SH01")
def live_ids(p) -> list[str]:
return [s.sid for s in p.sleeves if s.kind == "pairs" or s.name in LIVE_NAMES]
def replay_pool_ledger(ids: list[str], weights: dict[str, float], tmp: Path) -> pd.Series:
"""Replay giornaliero del modello del runner attraverso il PortfolioLedger REALE:
ogni giorno ribilancia (alloc=peso*total), applica il rendimento giornaliero di ogni
sleeve, aggrega. Ritorna la serie di equity totale (indicizzata per data)."""
eq = all_sleeve_equities()
rets = pd.DataFrame({i: eq[i].pct_change().fillna(0.0) for i in ids})
ledger = PortfolioLedger("VALIDATE", total_capital=1000.0, data_dir=tmp)
sleeve_cap = {i: weights[i] * ledger.total_capital for i in ids}
out = []
for day, row in rets.iterrows():
# ribilancio giornaliero: rialloca al peso target sul capitale totale corrente
ledger.total_capital = sum(sleeve_cap.values())
alloc = ledger.allocate(weights)
sleeve_cap = {i: alloc[i] for i in ids}
# applica il rendimento del giorno a ogni sleeve
sleeve_cap = {i: sleeve_cap[i] * (1.0 + row[i]) for i in ids}
ledger.update_equity(sleeve_cap)
out.append((day, ledger.equity))
return pd.Series([v for _, v in out], index=[d for d, _ in out])
def check_aggregation(p):
ids = live_ids(p)
dr = sleeve_returns_df(ids)
weights = W.weight_vector(p.weighting, ids, dr, weights=p.weights, caps=p.caps,
clusters={s.sid: (s.cluster or s.sid) for s in p.sleeves}, lookback=p.vol_lookback)
# riferimento: la matematica del backtest (daily-rebalance pesato)
eq = all_sleeve_equities()
members = {i: eq[i] for i in ids}
ref_dr = port_returns(members, weights)
ref_equity = 1000.0 * (1.0 + ref_dr).cumprod()
import tempfile, shutil
tmp = Path(tempfile.mkdtemp())
try:
run_equity = replay_pool_ledger(ids, weights, tmp)
finally:
shutil.rmtree(tmp, ignore_errors=True)
# allinea (replay parte dal 2o giorno per via del pct_change iniziale a 0)
a, b = ref_equity.align(run_equity, join="inner")
rel_err = float((a - b).abs().max() / a.abs().max())
end_ref, end_run = float(a.iloc[-1]), float(b.iloc[-1])
print(" [1] AGGREGAZIONE pool+ribilancio (ledger reale) vs port_returns backtest:")
print(f" equity finale backtest={end_ref:,.2f} runner-replay={end_run:,.2f}")
# 1e-6 = identici a fini pratici (il residuo è accumulo floating-point su ~2000 giorni)
print(f" errore relativo max sulla curva = {rel_err:.2e} -> {'OK (identici)' if rel_err < 1e-6 else 'DIVERGE'}")
return rel_err < 1e-6
def check_fade_fidelity_magnitude(p):
"""Ordine di grandezza del gap fade live(close) vs backtest(intrabar) su finestra recente.
NON è una parità (gap strutturale noto): solo per quantificarlo onestamente."""
from src.data.downloader import load_data
from scripts.analysis.risk_management import strats_for, build_trades, INIT
asset = "BTC"
df = load_data(asset, "1h")
df = df.iloc[-24 * 365:].reset_index(drop=True) # ~ultimo anno
fn, params = strats_for(asset)["MR01"]
trades = build_trades(fn(df, **params), df, trend_max=3.0)
bt_ret = 0.0
cap = INIT
for i, j, ret in sorted(trades, key=lambda t: t[1]):
cap = max(cap + cap * 0.15 * ret, 10.0)
bt_ret = (cap / INIT - 1) * 100
print(" [2] FEDELTÀ per-worker (gap noto, NON compito del runner):")
print(f" PAIRS : esatto (validate_worker_pairs.py)")
print(f" FADE : backtest intrabar MR01 {asset} ultimo anno = {bt_ret:+.1f}% "
f"(il live close-based diverge: vedi nota nel docstring)")
print(f" SHAPE : exit a tempo -> tick close coincide col backtest a meno del bar-timing")
def main():
p = PORTFOLIOS["PORT06"]
print("=" * 92)
print(" VALIDAZIONE PortfolioRunner — PORT06 (sleeve LIVE: fade+pairs+shape)")
print("=" * 92)
ok = check_aggregation(p)
print()
check_fade_fidelity_magnitude(p)
print()
print(" VERDETTO:")
print(f" livello POOL+RIBILANCIO+LEDGER del runner == backtest: {'CERTIFICATO' if ok else 'DA RIVEDERE'}")
print(" fedeltà per-worker: pairs esatta; fade approssimata (gap intrabar noto); shape a tempo ok")
if __name__ == "__main__":
main()
+11 -3
View File
@@ -9,12 +9,18 @@ sys.path.insert(0, str(PROJECT_ROOT))
from src.portfolio.base import Portfolio, SleeveSpec # noqa: E402
# Universo live tradabile (8 asset con feed Cerbero v2 + parquet). ROT02/TSM01 ci ruotano sopra.
UNIVERSE8 = ["ADA", "BNB", "BTC", "DOGE", "ETH", "LTC", "SOL", "XRP"]
FADE = [SleeveSpec(kind="single", name=c, sid=f"{c}_{a}", asset=a, cluster=f"{a}-rev")
for a in ("BTC", "ETH") for c in ("MR01", "MR02", "MR07")]
HONEST = [
# DIP01: single-asset 1h -> StrategyWorker (Strategy DIP01_dip_buy). TR01/ROT02: multi-asset.
SleeveSpec(kind="single", name="DIP01", sid="DIP01_BTC", asset="BTC", cluster="BTC-rev"),
SleeveSpec(kind="single", name="TR01", sid="TR01_basket", cluster="trend"),
SleeveSpec(kind="single", name="ROT02", sid="ROT02_rot", cluster="rotation"),
SleeveSpec(kind="basket", name="TR01", sid="TR01_basket", cluster="trend",
params={"universe": ["BNB", "BTC", "DOGE", "SOL", "XRP"], "tf": "4h"}),
SleeveSpec(kind="rotation", name="ROT02", sid="ROT02_rot", cluster="rotation",
params={"universe": UNIVERSE8, "tf": "1d", "top_k": 3, "gross": 0.45}),
]
PAIRS = [
SleeveSpec(kind="pairs", name="PR01", sid="PR_ETHBTC", a="ETH", b="BTC", cluster="ETH-rev"),
@@ -23,7 +29,9 @@ PAIRS = [
SleeveSpec(kind="pairs", name="PR01", sid="PR_BTCLTC", a="BTC", b="LTC", cluster="BTC-rev"),
SleeveSpec(kind="pairs", name="PR01", sid="PR_ETHSOL", a="ETH", b="SOL", cluster="ETH-rev"),
]
TSM = [SleeveSpec(kind="single", name="TSM01", sid="TSM01", cluster="trend")]
TSM = [SleeveSpec(kind="tsmom", name="TSM01", sid="TSM01", cluster="trend",
params={"universe": UNIVERSE8, "tf": "1d",
"horizons": [63, 126, 252], "thr": 1.0, "gross": 0.30})]
SHAPE = [SleeveSpec(kind="ml", name="SH01", sid=f"SH_{a}", asset=a, cluster="shape")
for a in ("BTC", "ETH")]
+54
View File
@@ -0,0 +1,54 @@
"""DIP01 — Dip-buy mean-reversion single-asset (z-score sotto-banda). Honest family.
Replica live della logica validata in scripts/analysis/honest_improve2.dip_market_gated
(con market_n=0, come lo sleeve DIP01_BTC del portafoglio): compra quando lo z-score del
prezzo rispetto a SMA(n) incrocia sotto -z_in; esce a TP=SMA, SL=close-sl_atr*ATR, o max_bars.
"""
from __future__ import annotations
import sys
from pathlib import Path
import numpy as np
import pandas as pd
PROJECT_ROOT = Path(__file__).resolve().parents[2]
sys.path.insert(0, str(PROJECT_ROOT))
from src.strategies.base import Strategy, Signal # noqa: E402
def _atr(df, n=14):
h, l, c = df["high"].values, df["low"].values, df["close"].values
pc = np.roll(c, 1); pc[0] = c[0]
tr = np.maximum(h - l, np.maximum(np.abs(h - pc), np.abs(l - pc)))
return pd.Series(tr).rolling(n).mean().values
class Dip01DipBuy(Strategy):
name = "DIP01_dip_buy"
description = "Dip-buy mean-reversion single-asset (z-score), exit TP=SMA/SL=ATR/max_bars"
default_assets = ["BTC"]
default_timeframes = ["1h"]
fee_rt = 0.001
leverage = 3.0
position_size = 0.15
def generate_signals(self, df: pd.DataFrame, ts: pd.DatetimeIndex,
n: int = 50, z_in: float = 2.5, sl_atr: float = 2.5,
max_bars: int = 24, **params) -> list[Signal]:
c = df["close"].values
ma = pd.Series(c).rolling(n).mean().values
sd = pd.Series(c).rolling(n).std().values
a = _atr(df, 14)
z = (c - ma) / np.where(sd == 0, np.nan, sd)
out: list[Signal] = []
for i in range(n + 14, len(c)):
if np.isnan(z[i]) or np.isnan(a[i]) or np.isnan(ma[i]):
continue
if z[i] <= -z_in and z[i - 1] > -z_in:
out.append(Signal(idx=i, direction=1, entry_price=float(c[i]),
metadata={"tp": float(ma[i]),
"sl": float(c[i] - sl_atr * a[i]),
"max_bars": int(max_bars)}))
return out
+87
View File
@@ -0,0 +1,87 @@
"""BasketTrendWorker (TR01): EMA20>EMA100 long/flat su un paniere, equal-weight.
Replica live di honest_improve2._tr_basket_daily."""
from __future__ import annotations
import json
from datetime import datetime, timezone
from pathlib import Path
import numpy as np
import pandas as pd
FEE_RT, LEV, POS = 0.001, 3.0, 0.15
def _ema(x, n):
return pd.Series(x).ewm(span=n, adjust=False).mean().values
class BasketTrendWorker:
def __init__(self, universe, tf="4h", capital=1000.0, position_size=POS,
leverage=LEV, fee_rt=FEE_RT, name="TR01_basket",
data_dir=Path("data/portfolio_paper")):
self.universe = list(universe)
self.tf = tf
self.initial_capital = capital
self.capital = capital
self.position_size = position_size
self.leverage = leverage
self.fee_rt = fee_rt
self.worker_id = f"{name}__{'-'.join(self.universe)}__{tf}"
self.work_dir = Path(data_dir) / self.worker_id
self.work_dir.mkdir(parents=True, exist_ok=True)
self.status_path = self.work_dir / "status.json"
self.trades_path = self.work_dir / "trades.jsonl"
self.positions = {a: 0.0 for a in self.universe}
self.last_bar_ts = {a: 0 for a in self.universe}
self.in_position = False
self._load()
def _load(self):
if self.status_path.exists():
s = json.loads(self.status_path.read_text())
self.capital = s.get("capital", self.capital)
self.positions = {**self.positions, **s.get("positions", {})}
self.last_bar_ts = {**self.last_bar_ts, **s.get("last_bar_ts", {})}
self.in_position = any(v > 0 for v in self.positions.values())
def _save(self):
self.status_path.write_text(json.dumps({
"capital": round(self.capital, 2), "positions": self.positions,
"last_bar_ts": self.last_bar_ts,
"ts": datetime.now(timezone.utc).isoformat()}, indent=2))
def tick(self, data: dict):
rets = []
for a in self.universe:
df = data.get(a)
if df is None or len(df) < 110:
continue
c = df["close"].values
ef, es = _ema(c, 20)[-1], _ema(c, 100)[-1]
target = 1.0 if ef > es else 0.0
bar_ts = int(df["timestamp"].iloc[-1])
prev = self.positions[a]
if self.last_bar_ts[a] and bar_ts > self.last_bar_ts[a] and prev > 0:
r = (c[-1] - c[-2]) / c[-2]
rets.append(self.position_size * self.leverage * r * prev)
if target != prev:
self.capital -= self.capital * self.position_size * (self.fee_rt / 2) * abs(target - prev) / len(self.universe)
self._log(a, prev, target, float(c[-1]))
self.positions[a] = target
self.last_bar_ts[a] = bar_ts
if rets:
self.capital = max(self.capital * (1 + float(np.mean(rets))), 10.0)
self.in_position = any(v > 0 for v in self.positions.values())
self._save()
def _log(self, asset, frm, to, price):
with open(self.trades_path, "a") as f:
f.write(json.dumps({"ts": datetime.now(timezone.utc).isoformat(),
"asset": asset, "from": frm, "to": to,
"price": round(price, 6), "capital": round(self.capital, 2)}) + "\n")
@property
def status_summary(self):
longs = [a for a, v in self.positions.items() if v > 0]
return f"{self.worker_id}: cap={self.capital:.0f} long={longs}"
+109
View File
@@ -0,0 +1,109 @@
"""RotationWorker (ROT02): dual-momentum top-k risk-gated, ribilancio giornaliero.
Replica live di honest_improve2._rot_daily_equity (lookback 60, top_k 3, gross 0.45, SMA100 gate)."""
from __future__ import annotations
import json
from datetime import datetime, timezone
from pathlib import Path
import numpy as np
import pandas as pd
FEE_RT = 0.001
def _panel(data: dict, universe: list):
"""Allinea {asset: df} sui timestamp comuni -> (df_panel, cols presenti)."""
frames = {}
for a in universe:
df = data.get(a)
if df is not None and len(df):
frames[a] = df[["timestamp", "close"]].rename(columns={"close": a})
if not frames:
return None, []
panel = None
for a, f in frames.items():
panel = f if panel is None else panel.merge(f, on="timestamp", how="inner")
panel = panel.sort_values("timestamp").reset_index(drop=True)
cols = [a for a in universe if a in frames]
return panel, cols
class RotationWorker:
def __init__(self, universe, lookback=60, top_k=3, gross=0.45, regime_n=100,
tf="1d", capital=1000.0, fee_rt=FEE_RT, name="ROT02_rot",
data_dir=Path("data/portfolio_paper")):
self.universe = list(universe)
self.lookback = lookback
self.top_k = top_k
self.gross = gross
self.regime_n = regime_n
self.tf = tf
self.initial_capital = capital
self.capital = capital
self.fee_rt = fee_rt
self.worker_id = f"{name}__{tf}"
self.work_dir = Path(data_dir) / self.worker_id
self.work_dir.mkdir(parents=True, exist_ok=True)
self.status_path = self.work_dir / "status.json"
self.trades_path = self.work_dir / "trades.jsonl"
self.weights = {a: 0.0 for a in self.universe}
self.last_bar_ts = 0
self.in_position = False
self._load()
def _load(self):
if self.status_path.exists():
s = json.loads(self.status_path.read_text())
self.capital = s.get("capital", self.capital)
self.weights = {**{a: 0.0 for a in self.universe}, **s.get("weights", {})}
self.last_bar_ts = s.get("last_bar_ts", 0)
self.in_position = any(v > 0 for v in self.weights.values())
def _save(self):
self.status_path.write_text(json.dumps({
"capital": round(self.capital, 2), "weights": self.weights,
"last_bar_ts": self.last_bar_ts,
"ts": datetime.now(timezone.utc).isoformat()}, indent=2))
def tick(self, data: dict):
panel, cols = _panel(data, self.universe)
if panel is None or len(panel) < max(self.lookback + 1, self.regime_n + 1) or "BTC" not in cols:
return
P = panel[cols].values
bar_ts = int(panel["timestamp"].iloc[-1])
# 1) realizza il rendimento dei pesi correnti sull'ultima barra chiusa
if self.last_bar_ts and bar_ts > self.last_bar_ts:
day_ret = P[-1] / P[-2] - 1.0
port_r = sum(self.weights.get(cols[k], 0.0) * day_ret[k] for k in range(len(cols)))
self.capital = max(self.capital * (1.0 + float(port_r)), 10.0)
# 2) ricalcola pesi target
btc = P[:, cols.index("BTC")]
bma = pd.Series(btc).rolling(self.regime_n).mean().values
risk_on = btc[-1] > bma[-1] if not np.isnan(bma[-1]) else False
mom = P[-1] / P[-1 - self.lookback] - 1.0
order = np.argsort(mom)[::-1]
chosen = [k for k in order if mom[k] > 0][: self.top_k] if risk_on else []
nw = {a: 0.0 for a in self.universe}
for k in chosen:
nw[cols[k]] = self.gross / len(chosen)
# 3) fee sul turnover
turnover = sum(abs(nw[a] - self.weights.get(a, 0.0)) for a in self.universe)
self.capital -= self.capital * turnover * (self.fee_rt / 2)
if turnover > 0:
self._log(nw, float(self.capital))
self.weights = nw
self.last_bar_ts = bar_ts
self.in_position = any(v > 0 for v in nw.values())
self._save()
def _log(self, weights, cap):
with open(self.trades_path, "a") as f:
f.write(json.dumps({"ts": datetime.now(timezone.utc).isoformat(),
"weights": {a: round(w, 4) for a, w in weights.items() if w > 0},
"capital": round(cap, 2)}) + "\n")
@property
def status_summary(self):
held = {a: round(w, 3) for a, w in self.weights.items() if w > 0}
return f"{self.worker_id}: cap={self.capital:.0f} held={held}"
+1
View File
@@ -17,6 +17,7 @@ _REGISTRY: dict[str, type[Strategy]] = {}
# scripts/waste/: l'edge storico era un artefatto di look-ahead
# (vedi scripts/analysis/oos_validation.py).
MODULE_MAP = {
"DIP01_dip_buy": ("DIP01_dip_buy", "Dip01DipBuy"),
"MR01_bollinger_fade": ("MR01_bollinger_fade", "BollingerFade"),
"MR02_donchian_fade": ("MR02_donchian_fade", "DonchianFade"),
"MR07_return_reversal": ("MR07_return_reversal", "ReturnReversal"),
+92
View File
@@ -0,0 +1,92 @@
"""TsmomWorker (TSM01): consenso TSMOM multi-orizzonte risk-gated, ribilancio giornaliero.
Replica live di tsmom_research.tsmom_sim (horizons 63/126/252, thr 1.0, gross 0.30, SMA100 gate)."""
from __future__ import annotations
import json
from datetime import datetime, timezone
from pathlib import Path
import numpy as np
import pandas as pd
from src.live.rotation_worker import _panel, FEE_RT
class TsmomWorker:
def __init__(self, universe, horizons=(63, 126, 252), thr=1.0, gross=0.30,
regime_n=100, tf="1d", capital=1000.0, fee_rt=FEE_RT,
name="TSM01", data_dir=Path("data/portfolio_paper")):
self.universe = list(universe)
self.horizons = tuple(horizons)
self.thr = thr
self.gross = gross
self.regime_n = regime_n
self.tf = tf
self.initial_capital = capital
self.capital = capital
self.fee_rt = fee_rt
self.worker_id = f"{name}__{tf}"
self.work_dir = Path(data_dir) / self.worker_id
self.work_dir.mkdir(parents=True, exist_ok=True)
self.status_path = self.work_dir / "status.json"
self.trades_path = self.work_dir / "trades.jsonl"
self.weights = {a: 0.0 for a in self.universe}
self.last_bar_ts = 0
self.in_position = False
self._load()
def _load(self):
if self.status_path.exists():
s = json.loads(self.status_path.read_text())
self.capital = s.get("capital", self.capital)
self.weights = {**{a: 0.0 for a in self.universe}, **s.get("weights", {})}
self.last_bar_ts = s.get("last_bar_ts", 0)
self.in_position = any(v > 0 for v in self.weights.values())
def _save(self):
self.status_path.write_text(json.dumps({
"capital": round(self.capital, 2), "weights": self.weights,
"last_bar_ts": self.last_bar_ts,
"ts": datetime.now(timezone.utc).isoformat()}, indent=2))
def tick(self, data: dict):
need = max(max(self.horizons) + 1, self.regime_n + 1)
panel, cols = _panel(data, self.universe)
if panel is None or len(panel) < need or "BTC" not in cols:
return
P = panel[cols].values
bar_ts = int(panel["timestamp"].iloc[-1])
if self.last_bar_ts and bar_ts > self.last_bar_ts:
day_ret = P[-1] / P[-2] - 1.0
port_r = sum(self.weights.get(cols[k], 0.0) * day_ret[k] for k in range(len(cols)))
self.capital = max(self.capital * (1.0 + float(port_r)), 10.0)
btc = P[:, cols.index("BTC")]
bma = pd.Series(btc).rolling(self.regime_n).mean().values
risk_on = btc[-1] > bma[-1] if not np.isnan(bma[-1]) else False
score = np.zeros(len(cols))
for h in self.horizons:
score += np.sign(P[-1] / P[-1 - h] - 1.0)
score /= len(self.horizons)
chosen = [k for k in range(len(cols)) if score[k] >= self.thr] if risk_on else []
nw = {a: 0.0 for a in self.universe}
for k in chosen:
nw[cols[k]] = self.gross / len(chosen)
turnover = sum(abs(nw[a] - self.weights.get(a, 0.0)) for a in self.universe)
self.capital -= self.capital * turnover * (self.fee_rt / 2)
if turnover > 0:
self._log(nw, float(self.capital))
self.weights = nw
self.last_bar_ts = bar_ts
self.in_position = any(v > 0 for v in nw.values())
self._save()
def _log(self, weights, cap):
with open(self.trades_path, "a") as f:
f.write(json.dumps({"ts": datetime.now(timezone.utc).isoformat(),
"weights": {a: round(w, 4) for a, w in weights.items() if w > 0},
"capital": round(cap, 2)}) + "\n")
@property
def status_summary(self):
held = {a: round(w, 3) for a, w in self.weights.items() if w > 0}
return f"{self.worker_id}: cap={self.capital:.0f} held={held}"
+97 -33
View File
@@ -1,24 +1,42 @@
"""PortfolioRunner: faccia live del portafoglio (capitale pool, sizing, ribilancio, ledger).
Riusa i worker esistenti come esecutori e il data layer Cerbero v2."""
Riusa i worker esistenti come esecutori e il data layer Cerbero v2.
Worker per tipo di sleeve:
single (fade/dip) -> StrategyWorker | ml (shape) -> MLWorkerWrapper(StrategyWorker)
pairs -> PairsWorker (2 gambe) | basket (TR01) -> BasketTrendWorker
rotation (ROT02) -> RotationWorker | tsmom (TSM01) -> TsmomWorker
Feed: il runner fetcha candele 1h da Cerbero v2 e le RESAMPLA a 4h/1d (come get_df nel
backtest) per i worker a cadenza piu' lenta. Il lookback per asset e' dimensionato sul
worker piu' esigente (TSM01 usa 252 giorni)."""
from __future__ import annotations
from pathlib import Path
import pandas as pd
from src.portfolio.base import SleeveSpec, Portfolio
from src.portfolio.ledger import PortfolioLedger
from src.live.strategy_worker import StrategyWorker
from src.live.pairs_worker import PairsWorker
from src.live.basket_trend_worker import BasketTrendWorker
from src.live.rotation_worker import RotationWorker
from src.live.tsmom_worker import TsmomWorker
from src.live.multi_runner import MLWorkerWrapper
from src.live.strategy_loader import load_strategy
# Codice-breve sleeve -> nome modulo Strategy in scripts/strategies/
# Codice-breve sleeve -> nome modulo Strategy in scripts/strategies/ (worker single/ml)
_STRAT_MODULE = {
"MR01": "MR01_bollinger_fade", "MR02": "MR02_donchian_fade",
"MR07": "MR07_return_reversal", "SH01": "SH01_shape_ml",
# DIP01/TR01/ROT02 sono honest a sé: vedi nota nel design (worker dedicati in fase 2)
"DIP01": "DIP01_dip_buy",
}
_MULTI_KINDS = ("basket", "rotation", "tsmom")
DATA_DIR = Path("data/portfolio_paper")
# giorni di storia da fetchare per timeframe (TSM01 1d usa 252 barre -> ~440 giorni col buffer)
_LOOKBACK_DAYS = {"1h": 90, "4h": 220, "1d": 440}
def build_worker_for(spec: SleeveSpec, alloc_capital: float, leverage: float,
data_dir: Path = DATA_DIR, position_size: float = 0.15):
@@ -29,10 +47,28 @@ def build_worker_for(spec: SleeveSpec, alloc_capital: float, leverage: float,
capital=alloc_capital, position_size=position_size, leverage=leverage,
fee_rt=0.001, name="PR01_pairs_reversion", data_dir=data_dir,
)
if spec.kind == "basket":
pr = spec.params
return BasketTrendWorker(
universe=pr["universe"], tf=pr.get("tf", "4h"), capital=alloc_capital,
position_size=position_size, leverage=leverage, data_dir=data_dir,
)
if spec.kind == "rotation":
pr = spec.params
return RotationWorker(
universe=pr["universe"], top_k=pr.get("top_k", 3), gross=pr.get("gross", 0.45),
tf=pr.get("tf", "1d"), capital=alloc_capital, data_dir=data_dir,
)
if spec.kind == "tsmom":
pr = spec.params
return TsmomWorker(
universe=pr["universe"], horizons=tuple(pr.get("horizons", (63, 126, 252))),
thr=pr.get("thr", 1.0), gross=pr.get("gross", 0.30),
tf=pr.get("tf", "1d"), capital=alloc_capital, data_dir=data_dir,
)
module = _STRAT_MODULE.get(spec.name)
if module is None:
raise ValueError(f"sleeve live non ancora supportato: {spec.name} "
f"(honest DIP01/TR01/ROT02 richiedono worker dedicati, fase 2)")
raise ValueError(f"sleeve live non supportato: {spec.name} (kind={spec.kind})")
strategy = load_strategy(module)
worker = StrategyWorker(
strategy=strategy, asset=spec.asset, tf=spec.tf, capital=alloc_capital,
@@ -63,13 +99,36 @@ def rebalance_allocations(ledger: PortfolioLedger, workers: dict, weights: dict[
ledger.save()
def _resample(df: pd.DataFrame, tf: str) -> pd.DataFrame:
"""Resampla candele 1h -> 4h/1d mantenendo timestamp ms reale (come get_df del backtest)."""
if tf == "1h":
return df
rule = {"4h": "4h", "1d": "1D"}[tf]
d = df.copy()
d["dt"] = pd.to_datetime(d["timestamp"], unit="ms", utc=True)
d = d.set_index("dt")
agg = d.resample(rule).agg({"open": "first", "high": "max", "low": "min",
"close": "last", "volume": "sum"}).dropna()
epoch = pd.Timestamp("1970-01-01", tz="UTC")
agg["timestamp"] = ((agg.index - epoch) // pd.Timedelta(milliseconds=1)).astype("int64")
return agg.reset_index(drop=True)
def _spec_assets_tf(spec: SleeveSpec):
"""(lista asset, tf) coinvolti da uno sleeve."""
if spec.kind == "pairs":
return [spec.a, spec.b], spec.tf
if spec.kind in _MULTI_KINDS:
return list(spec.params["universe"]), spec.params.get("tf", "1d" if spec.kind != "basket" else "4h")
return [spec.asset], spec.tf
def run(config_path: str = "portfolios.yml"):
"""Loop live a portafoglio. Data layer Cerbero v2; ribilancio a fine giornata UTC.
Gli sleeve senza worker live (honest DIP01/TR01/ROT02) vengono SALTATI con warning
(restano solo in backtest); i pesi sono rinormalizzati sugli sleeve eseguibili."""
"""Loop live a portafoglio (tutti i tipi di sleeve). Data layer Cerbero v2 con resample;
ribilancio a cambio giornata UTC."""
import time
from datetime import datetime, timezone, timedelta
import pandas as pd
import yaml
from src.portfolio.base import load_active_portfolio
from src.portfolio.sleeves import sleeve_returns_df
from src.portfolio import weighting as W
@@ -77,13 +136,11 @@ def run(config_path: str = "portfolios.yml"):
from src.live.multi_runner import INSTRUMENT_MAP
p: Portfolio = load_active_portfolio(config_path)
import yaml as _yaml
_ov = (_yaml.safe_load(__import__("pathlib").Path(config_path).read_text()) or {}).get("overrides", {})
_ov = (yaml.safe_load(Path(config_path).read_text()) or {}).get("overrides", {})
poll = int(_ov.get("poll_seconds", 60))
def _supported(s):
return s.kind == "pairs" or s.name in _STRAT_MODULE
return s.kind in ("pairs",) + _MULTI_KINDS or s.name in _STRAT_MODULE
live_specs = [s for s in p.sleeves if _supported(s)]
skipped = [s.sid for s in p.sleeves if not _supported(s)]
if skipped:
@@ -100,40 +157,47 @@ def run(config_path: str = "portfolios.yml"):
alloc = ledger.allocate(weights)
workers = {s.sid: build_worker_for(s, alloc[s.sid], p.leverage) for s in live_specs}
# lookback (giorni) richiesto per ogni asset = max sui worker che lo usano
asset_days: dict[str, int] = {}
for s in live_specs:
assets, tf = _spec_assets_tf(s)
for a in assets:
asset_days[a] = max(asset_days.get(a, 0), _LOOKBACK_DAYS.get(tf, 90))
inst_map = dict(INSTRUMENT_MAP)
last_day = ""
while True:
try:
keys = set()
for s in live_specs:
if s.kind == "pairs":
keys.add((s.a, s.tf)); keys.add((s.b, s.tf))
else:
keys.add((s.asset, s.tf))
cache = {}
end = datetime.now(timezone.utc); start = end - timedelta(days=60)
for asset, tf in keys:
# fetch 1h per asset al lookback massimo richiesto
raw1h: dict[str, pd.DataFrame] = {}
end = datetime.now(timezone.utc)
for asset, days in asset_days.items():
inst = inst_map.get(asset, f"{asset}-PERPETUAL")
start = end - timedelta(days=days)
candles = client.get_historical_v2(inst, start.strftime("%Y-%m-%d"),
end.strftime("%Y-%m-%d"), tf)
end.strftime("%Y-%m-%d"), "1h")
if candles:
df = pd.DataFrame(candles)
df["timestamp"] = df["timestamp"].astype("int64")
cache[(asset, tf)] = df.sort_values("timestamp").reset_index(drop=True)
raw1h[asset] = df.sort_values("timestamp").reset_index(drop=True)
# tick di ogni worker col suo timeframe (resample dal 1h)
for s in live_specs:
w = workers[s.sid]
assets, tf = _spec_assets_tf(s)
if any(a not in raw1h for a in assets):
continue
res = {a: _resample(raw1h[a], tf) for a in assets}
if s.kind == "pairs":
ka, kb = (s.a, s.tf), (s.b, s.tf)
if ka in cache and kb in cache:
w.tick(cache[ka], cache[kb])
w.tick(res[s.a], res[s.b])
elif s.kind in _MULTI_KINDS:
w.tick(res)
else:
key = (s.asset, s.tf)
if key in cache:
inner = getattr(w, "worker", w)
if hasattr(w, "needs_training") and w.needs_training():
w.train(cache[key], hold=inner.hold_bars)
w.tick(cache[key])
df = res[s.asset]
inner = getattr(w, "worker", w)
if hasattr(w, "needs_training") and w.needs_training():
w.train(df, hold=inner.hold_bars)
w.tick(df)
ledger.update_equity({sid: _worker_equity(wk) for sid, wk in workers.items()})
+30
View File
@@ -0,0 +1,30 @@
import numpy as np
import pandas as pd
from src.live.basket_trend_worker import BasketTrendWorker
def _ramp_df(n=300, slope=1.0):
c = np.linspace(100, 100 + slope * n, n)
ts = (pd.date_range("2024-01-01", periods=n, freq="4h", tz="UTC").astype("int64") // 10**6)
return pd.DataFrame({"timestamp": ts, "open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def test_basket_goes_long_in_uptrend(tmp_path):
w = BasketTrendWorker(universe=["AAA", "BBB"], tf="4h", capital=1000.0, data_dir=tmp_path)
data = {"AAA": _ramp_df(slope=1.0), "BBB": _ramp_df(slope=1.0)}
w.tick(data)
assert w.positions["AAA"] == 1.0 and w.positions["BBB"] == 1.0
def test_basket_flat_in_downtrend(tmp_path):
w = BasketTrendWorker(universe=["AAA"], tf="4h", capital=1000.0, data_dir=tmp_path)
data = {"AAA": _ramp_df(slope=-1.0)}
w.tick(data)
assert w.positions["AAA"] == 0.0
def test_basket_persists_and_resumes(tmp_path):
w = BasketTrendWorker(universe=["AAA"], tf="4h", capital=1000.0, data_dir=tmp_path)
w.tick({"AAA": _ramp_df(slope=1.0)})
w2 = BasketTrendWorker(universe=["AAA"], tf="4h", capital=1000.0, data_dir=tmp_path)
assert w2.positions["AAA"] == 1.0 # stato ripreso da status.json
+13
View File
@@ -0,0 +1,13 @@
import pandas as pd
from src.data.downloader import load_data
from scripts.strategies.DIP01_dip_buy import Dip01DipBuy
def test_dip01_generates_long_signals_with_exits():
df = load_data("BTC", "1h").iloc[-5000:].reset_index(drop=True)
ts = pd.to_datetime(df["timestamp"], unit="ms", utc=True)
sigs = Dip01DipBuy().generate_signals(df, ts, asset="BTC", tf="1h")
assert len(sigs) > 0
s = sigs[0]
assert s.direction == 1
assert {"tp", "sl", "max_bars"} <= set(s.metadata)
+32
View File
@@ -0,0 +1,32 @@
import numpy as np
import pandas as pd
from src.live.rotation_worker import RotationWorker
def _df(n=200, slope=1.0):
c = np.linspace(100, 100 + slope * n, n)
ts = (pd.date_range("2023-01-01", periods=n, freq="1D", tz="UTC").astype("int64") // 10**6)
return pd.DataFrame({"timestamp": ts, "open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def test_rotation_picks_top_momentum_when_risk_on(tmp_path):
w = RotationWorker(universe=["BTC", "AAA", "BBB"], top_k=2, gross=0.45, data_dir=tmp_path)
data = {"BTC": _df(slope=1.0), "AAA": _df(slope=3.0), "BBB": _df(slope=0.1)}
w.tick(data)
assert w.weights["AAA"] > 0
assert abs(sum(w.weights.values()) - 0.45) < 1e-9
def test_rotation_flat_when_risk_off(tmp_path):
# BTC in downtrend -> risk_off -> nessuna posizione
w = RotationWorker(universe=["BTC", "AAA"], top_k=1, gross=0.45, data_dir=tmp_path)
data = {"BTC": _df(slope=-1.0), "AAA": _df(slope=3.0)}
w.tick(data)
assert sum(w.weights.values()) == 0.0
def test_rotation_persists_and_resumes(tmp_path):
w = RotationWorker(universe=["BTC", "AAA"], top_k=1, gross=0.45, data_dir=tmp_path)
w.tick({"BTC": _df(slope=1.0), "AAA": _df(slope=3.0)})
w2 = RotationWorker(universe=["BTC", "AAA"], top_k=1, gross=0.45, data_dir=tmp_path)
assert w2.weights == w.weights
+39
View File
@@ -0,0 +1,39 @@
"""T5: integrazione worker honest/TSM01 nel PortfolioRunner."""
from src.portfolio.runner import build_worker_for, _STRAT_MODULE, _MULTI_KINDS
from src.portfolio.base import SleeveSpec
from src.live.basket_trend_worker import BasketTrendWorker
from src.live.rotation_worker import RotationWorker
from src.live.tsmom_worker import TsmomWorker
from src.live.strategy_worker import StrategyWorker
from scripts.portfolios._defs import PORTFOLIOS
def test_build_basket_worker(tmp_path):
spec = SleeveSpec(kind="basket", name="TR01", sid="TR01_basket",
params={"universe": ["BNB", "BTC"], "tf": "4h"})
w = build_worker_for(spec, alloc_capital=120.0, leverage=2.0, data_dir=tmp_path)
assert isinstance(w, BasketTrendWorker) and w.capital == 120.0
def test_build_rotation_and_tsmom(tmp_path):
rot = SleeveSpec(kind="rotation", name="ROT02", sid="ROT02_rot",
params={"universe": ["BTC", "ETH"], "tf": "1d", "top_k": 1, "gross": 0.45})
tsm = SleeveSpec(kind="tsmom", name="TSM01", sid="TSM01",
params={"universe": ["BTC", "ETH"], "tf": "1d", "gross": 0.30})
wr = build_worker_for(rot, 100.0, 2.0, data_dir=tmp_path)
wt = build_worker_for(tsm, 100.0, 2.0, data_dir=tmp_path)
assert isinstance(wr, RotationWorker) and wr.capital == 100.0
assert isinstance(wt, TsmomWorker) and wt.capital == 100.0
def test_dip01_builds_as_strategy_worker(tmp_path):
spec = SleeveSpec(kind="single", name="DIP01", sid="DIP01_BTC", asset="BTC")
w = build_worker_for(spec, 80.0, 2.0, data_dir=tmp_path)
assert isinstance(w, StrategyWorker) and w.capital == 80.0
def test_port06_has_no_unsupported_sleeves():
p = PORTFOLIOS["PORT06"]
unsupported = [s.sid for s in p.sleeves
if not (s.kind in ("pairs",) + _MULTI_KINDS or s.name in _STRAT_MODULE)]
assert unsupported == []
+33
View File
@@ -0,0 +1,33 @@
import numpy as np
import pandas as pd
from src.live.tsmom_worker import TsmomWorker
def _df(n=300, slope=1.0):
c = np.linspace(100, 100 + slope * n, n)
ts = (pd.date_range("2023-01-01", periods=n, freq="1D", tz="UTC").astype("int64") // 10**6)
return pd.DataFrame({"timestamp": ts, "open": c, "high": c, "low": c, "close": c, "volume": 1.0})
def test_tsmom_selects_full_consensus_uptrend(tmp_path):
# tutti gli orizzonti positivi -> score=1>=thr; BTC su -> risk_on
w = TsmomWorker(universe=["BTC", "AAA"], horizons=(63, 126, 252), thr=1.0,
gross=0.30, data_dir=tmp_path)
data = {"BTC": _df(slope=1.0), "AAA": _df(slope=2.0)}
w.tick(data)
assert w.weights["BTC"] > 0 and w.weights["AAA"] > 0
assert abs(sum(w.weights.values()) - 0.30) < 1e-9
def test_tsmom_flat_when_risk_off(tmp_path):
w = TsmomWorker(universe=["BTC", "AAA"], thr=1.0, gross=0.30, data_dir=tmp_path)
data = {"BTC": _df(slope=-1.0), "AAA": _df(slope=2.0)}
w.tick(data)
assert sum(w.weights.values()) == 0.0
def test_tsmom_persists_and_resumes(tmp_path):
w = TsmomWorker(universe=["BTC", "AAA"], gross=0.30, data_dir=tmp_path)
w.tick({"BTC": _df(slope=1.0), "AAA": _df(slope=2.0)})
w2 = TsmomWorker(universe=["BTC", "AAA"], gross=0.30, data_dir=tmp_path)
assert w2.weights == w.weights