Spaces:
Sleeping
Sleeping
File size: 14,950 Bytes
e6021a3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 | """
Internal Pine Script Backtester.
Translates the *intent* of Pine Script strategies into Python
and executes them against real historical data via yfinance.
This is NOT a full Pine Script parser — it maps the generated
template strategies to the existing BacktestEngine.
Produces TradingView-style performance metrics:
- Net Profit, Gross Profit, Gross Loss
- Max Drawdown, Profit Factor
- Sharpe Ratio, Sortino Ratio
- Win Rate, Average Win/Loss
- Equity Curve, Monthly Returns
"""
from __future__ import annotations
import logging
import re
from typing import Any, Dict, List, Optional, Tuple
import numpy as np
import pandas as pd
logger = logging.getLogger(__name__)
class PineScriptBacktester:
"""
Simulates Pine Script strategies using Python.
For template-generated strategies, maps the strategy logic
directly to pandas operations for accurate backtesting.
"""
async def backtest(
self,
code: str,
ticker: str = "SPY",
period: str = "3y",
initial_capital: float = 100_000,
commission_pct: float = 0.1,
) -> Dict[str, Any]:
"""
Run a backtest on Pine Script code.
"""
from app.services.data_ingestion.yahoo import yahoo_adapter
# Fetch data
df = await yahoo_adapter.get_price_dataframe(ticker, period=period)
if df.empty or len(df) < 50:
raise ValueError(f"Insufficient data for {ticker}")
# Detect strategy type from code and run appropriate backtest
strategy_type = self._detect_strategy_type(code)
params = self._extract_parameters(code)
signals = self._generate_signals(df, strategy_type, params)
results = self._simulate_trades(df, signals, initial_capital, commission_pct)
return {
"ticker": ticker,
"period": period,
"strategy_type": strategy_type,
"initial_capital": initial_capital,
**results,
}
def _detect_strategy_type(self, code: str) -> str:
"""Determine strategy type from Pine Script code keywords."""
code_lower = code.lower()
if "ta.supertrend" in code_lower:
return "supertrend"
if "ta.macd" in code_lower:
return "macd"
if "ta.rsi" in code_lower and "ta.stoch" in code_lower:
return "stochastic_rsi"
if "ta.rsi" in code_lower:
return "rsi"
if "ta.bb" in code_lower or "bollinger" in code_lower or "ta.stdev" in code_lower:
return "bollinger"
if "ta.vwap" in code_lower or "vwap" in code_lower:
return "vwap"
if "ichimoku" in code_lower or "donchian" in code_lower:
return "ichimoku"
if "z_score" in code_lower or "z-score" in code_lower:
return "zscore"
if "ema" in code_lower and code_lower.count("ta.ema") >= 3:
return "ema_ribbon"
if "crossover" in code_lower and "sma" in code_lower:
return "sma_crossover"
if "ta.atr" in code_lower and "trail" in code_lower:
return "atr_trailing"
if "request.security" in code_lower:
return "multi_timeframe"
return "sma_crossover" # default
def _extract_parameters(self, code: str) -> Dict[str, Any]:
"""Extract input parameters from Pine Script code."""
params: Dict[str, Any] = {}
# Match input.int(value, ...) and input.float(value, ...)
int_matches = re.findall(r'(\w+)\s*=\s*input\.int\((\d+)', code)
float_matches = re.findall(r'(\w+)\s*=\s*input\.float\(([\d.]+)', code)
for name, val in int_matches:
params[name] = int(val)
for name, val in float_matches:
params[name] = float(val)
return params
def _generate_signals(
self,
df: pd.DataFrame,
strategy_type: str,
params: Dict[str, Any],
) -> pd.Series:
"""
Generate trading signals: +1 = buy, -1 = sell, 0 = hold.
Maps each strategy type to Python logic.
"""
close = df["Close"]
high = df["High"]
low = df["Low"]
volume = df["Volume"] if "Volume" in df.columns else pd.Series(0, index=df.index)
signals = pd.Series(0, index=df.index)
if strategy_type == "sma_crossover":
fast = params.get("fast_length", 20)
slow = params.get("slow_length", 50)
sma_fast = close.rolling(fast).mean()
sma_slow = close.rolling(slow).mean()
signals[sma_fast > sma_slow] = 1
signals[sma_fast <= sma_slow] = -1
elif strategy_type == "rsi":
length = params.get("rsi_length", 14)
oversold = params.get("oversold_level", 30)
overbought = params.get("overbought_level", 70)
delta = close.diff()
gain = delta.where(delta > 0, 0).rolling(length).mean()
loss = (-delta.where(delta < 0, 0)).rolling(length).mean()
rs = gain / (loss + 1e-10)
rsi = 100 - (100 / (1 + rs))
signals[rsi < oversold] = 1
signals[rsi > overbought] = -1
elif strategy_type == "macd":
fast = params.get("fast_len", 12)
slow = params.get("slow_len", 26)
sig = params.get("sig_len", 9)
ema_fast = close.ewm(span=fast).mean()
ema_slow = close.ewm(span=slow).mean()
macd_line = ema_fast - ema_slow
signal_line = macd_line.ewm(span=sig).mean()
hist = macd_line - signal_line
signals[(macd_line > signal_line) & (hist > 0)] = 1
signals[macd_line < signal_line] = -1
elif strategy_type == "bollinger":
length = params.get("length", 20)
mult = params.get("mult", 2.0)
basis = close.rolling(length).mean()
std = close.rolling(length).std()
upper = basis + mult * std
signals[close > upper] = 1
signals[close < basis] = -1
elif strategy_type == "supertrend":
atr_len = params.get("atr_len", 10)
factor = params.get("factor", 3.0)
tr = pd.concat([
high - low,
abs(high - close.shift(1)),
abs(low - close.shift(1))
], axis=1).max(axis=1)
atr = tr.rolling(atr_len).mean()
upper_band = (high + low) / 2 + factor * atr
lower_band = (high + low) / 2 - factor * atr
supertrend = pd.Series(0.0, index=df.index)
direction = pd.Series(1, index=df.index)
for i in range(1, len(df)):
if close.iloc[i] > upper_band.iloc[i-1]:
direction.iloc[i] = 1
elif close.iloc[i] < lower_band.iloc[i-1]:
direction.iloc[i] = -1
else:
direction.iloc[i] = direction.iloc[i-1]
signals[direction == 1] = 1
signals[direction == -1] = -1
elif strategy_type == "ema_ribbon":
emas = [8, 13, 21, 34, 55]
ema_vals = [close.ewm(span=e).mean() for e in emas]
bullish = all(ema_vals[i].iloc[-1] > ema_vals[i+1].iloc[-1] for i in range(len(ema_vals)-1))
for i in range(len(df)):
if all(ema_vals[j].iloc[i] > ema_vals[j+1].iloc[i] for j in range(len(ema_vals)-1) if i < len(ema_vals[j])):
signals.iloc[i] = 1
elif all(ema_vals[j].iloc[i] < ema_vals[j+1].iloc[i] for j in range(len(ema_vals)-1) if i < len(ema_vals[j])):
signals.iloc[i] = -1
elif strategy_type == "zscore":
lookback = params.get("lookback", 50)
entry_z = params.get("entry_z", 2.0)
mean = close.rolling(lookback).mean()
std = close.rolling(lookback).std()
z = (close - mean) / (std + 1e-10)
signals[z < -entry_z] = 1
signals[z > entry_z] = -1
else:
# Default: SMA crossover
sma20 = close.rolling(20).mean()
sma50 = close.rolling(50).mean()
signals[sma20 > sma50] = 1
signals[sma20 <= sma50] = -1
return signals
def _simulate_trades(
self,
df: pd.DataFrame,
signals: pd.Series,
initial_capital: float,
commission_pct: float,
) -> Dict[str, Any]:
"""
Simulate trading based on signals and compute all metrics.
"""
close = df["Close"].values
sig = signals.values
n = len(close)
# Track positions and equity
position = 0 # 0 or 1
entry_price = 0.0
capital = initial_capital
equity_curve = [initial_capital]
trades: List[Dict[str, Any]] = []
daily_returns: List[float] = [0.0]
for i in range(1, n):
if sig[i] == 1 and position == 0:
# Buy
shares = capital / close[i]
commission = capital * (commission_pct / 100)
capital -= commission
entry_price = close[i]
position = 1
trades.append({
"type": "ENTRY",
"date": str(df.index[i].date()) if hasattr(df.index[i], 'date') else str(df.index[i]),
"price": round(close[i], 2),
"shares": round(shares, 4),
})
elif sig[i] == -1 and position == 1:
# Sell
ret = (close[i] - entry_price) / entry_price
commission = capital * (1 + ret) * (commission_pct / 100)
capital = capital * (1 + ret) - commission
position = 0
trades.append({
"type": "EXIT",
"date": str(df.index[i].date()) if hasattr(df.index[i], 'date') else str(df.index[i]),
"price": round(close[i], 2),
"pnl_pct": round(ret * 100, 2),
"pnl_abs": round(capital - equity_curve[-1], 2),
})
# Update equity
if position == 1:
current_equity = capital * (close[i] / entry_price) if entry_price > 0 else capital
else:
current_equity = capital
daily_ret = (current_equity - equity_curve[-1]) / equity_curve[-1] if equity_curve[-1] > 0 else 0
daily_returns.append(daily_ret)
equity_curve.append(current_equity)
# Close open position
if position == 1:
ret = (close[-1] - entry_price) / entry_price
capital = capital * (1 + ret)
# Compute metrics
eq = np.array(equity_curve)
rets = np.array(daily_returns)
peak = np.maximum.accumulate(eq)
drawdown = (eq - peak) / (peak + 1e-10)
# Trade stats
pnl_list = [t.get("pnl_pct", 0) for t in trades if t["type"] == "EXIT"]
wins = [p for p in pnl_list if p > 0]
losses = [p for p in pnl_list if p < 0]
total_trades = len(pnl_list)
win_rate = len(wins) / total_trades if total_trades > 0 else 0
avg_win = np.mean(wins) if wins else 0
avg_loss = np.mean(losses) if losses else 0
net_profit = eq[-1] - initial_capital
net_profit_pct = (eq[-1] / initial_capital - 1) * 100
gross_profit = sum(p for p in pnl_list if p > 0)
gross_loss = abs(sum(p for p in pnl_list if p < 0))
profit_factor = gross_profit / (gross_loss + 1e-10) if gross_loss > 0 else float("inf")
max_drawdown = float(np.min(drawdown)) * 100
# Annualized metrics
trading_days = len(rets) - 1
ann_factor = 252 / max(trading_days, 1)
ann_return = ((eq[-1] / initial_capital) ** ann_factor - 1) * 100 if trading_days > 0 else 0
daily_std = np.std(rets[1:]) if len(rets) > 1 else 0
sharpe = (np.mean(rets[1:]) / (daily_std + 1e-10)) * np.sqrt(252) if daily_std > 0 else 0
downside = rets[rets < 0]
downside_std = np.std(downside) if len(downside) > 0 else 0
sortino = (np.mean(rets[1:]) / (downside_std + 1e-10)) * np.sqrt(252) if downside_std > 0 else 0
# Monthly returns
dates = df.index
monthly_returns = {}
month_start_eq = equity_curve[0]
for i in range(1, len(equity_curve)):
if i < len(dates):
month_key = dates[i].strftime("%Y-%m") if hasattr(dates[i], 'strftime') else str(dates[i])[:7]
if i > 0 and (i == len(equity_curve) - 1 or
(i < len(dates) and hasattr(dates[i], 'month') and
(dates[i].month != dates[i-1].month if i > 0 and hasattr(dates[i-1], 'month') else False))):
prev_key = dates[i-1].strftime("%Y-%m") if hasattr(dates[i-1], 'strftime') else str(dates[i-1])[:7]
monthly_returns[prev_key] = round((equity_curve[i] / month_start_eq - 1) * 100, 2) if month_start_eq > 0 else 0
month_start_eq = equity_curve[i]
# Downsample equity curve for response size
eq_len = len(equity_curve)
step = max(1, eq_len // 250)
sampled_equity = [
{"index": i, "equity": round(equity_curve[i], 2)}
for i in range(0, eq_len, step)
]
return {
"net_profit": round(net_profit, 2),
"net_profit_pct": round(net_profit_pct, 2),
"gross_profit_pct": round(gross_profit, 2),
"gross_loss_pct": round(gross_loss, 2),
"profit_factor": round(profit_factor, 4),
"max_drawdown_pct": round(max_drawdown, 2),
"sharpe_ratio": round(sharpe, 4),
"sortino_ratio": round(sortino, 4),
"annualized_return_pct": round(ann_return, 2),
"total_trades": total_trades,
"win_rate": round(win_rate, 4),
"avg_win_pct": round(avg_win, 2),
"avg_loss_pct": round(avg_loss, 2),
"max_consecutive_wins": _max_consecutive(pnl_list, positive=True),
"max_consecutive_losses": _max_consecutive(pnl_list, positive=False),
"equity_curve": sampled_equity,
"trades": trades[-50:], # last 50 trades
"monthly_returns": monthly_returns,
"final_equity": round(eq[-1], 2),
"trading_days": trading_days,
}
def _max_consecutive(values: list, positive: bool = True) -> int:
"""Count max consecutive wins or losses."""
max_count = 0
current = 0
for v in values:
if (positive and v > 0) or (not positive and v < 0):
current += 1
max_count = max(max_count, current)
else:
current = 0
return max_count
# Module singleton
pine_backtester = PineScriptBacktester()
|