Alice — an open-source AI trading bot covering equities, crypto, commodities, forex, and macro

master
py 117 lines 3.1 KB
Raw
1 """Paper trading engine."""
2
3 from __future__ import annotations
4
5 from dataclasses import dataclass, field
6 from datetime import datetime
7
8 import pandas as pd
9
10 from open_alice.strategies.base import Strategy
11
12
13 @dataclass
14 class PaperPosition:
15 symbol: str
16 shares: int
17 entry_price: float
18 entry_time: datetime
19
20
21 @dataclass
22 class Trade:
23 time: datetime
24 action: str
25 symbol: str
26 shares: int
27 price: float
28 pnl: float | None = None
29
30
31 @dataclass
32 class PaperPortfolio:
33 cash: float
34 positions: dict[str, PaperPosition] = field(default_factory=dict)
35 trades: list[Trade] = field(default_factory=list)
36
37 @property
38 def equity(self) -> float:
39 total = self.cash
40 for pos in self.positions.values():
41 total += pos.shares * pos.entry_price
42 return total
43
44 def buy(self, symbol: str, price: float, shares: int) -> None:
45 cost = shares * price
46 if cost > self.cash:
47 return
48 self.cash -= cost
49 self.positions[symbol] = PaperPosition(symbol, shares, price, datetime.now())
50 self.trades.append(Trade(
51 time=datetime.now(),
52 action="BUY",
53 symbol=symbol,
54 shares=shares,
55 price=price,
56 ))
57
58 def sell(self, symbol: str, price: float) -> None:
59 if symbol not in self.positions:
60 return
61 pos = self.positions.pop(symbol)
62 proceeds = pos.shares * price
63 self.cash += proceeds
64 pnl = proceeds - (pos.shares * pos.entry_price)
65 self.trades.append(Trade(
66 time=datetime.now(),
67 action="SELL",
68 symbol=symbol,
69 shares=pos.shares,
70 price=price,
71 pnl=pnl,
72 ))
73
74
75 def run_paper(
76 df: pd.DataFrame,
77 strategy: Strategy,
78 symbol: str,
79 initial_capital: float = 100_000.0,
80 risk_per_trade: float = 0.02,
81 ) -> PaperPortfolio:
82 """Run a paper trading simulation.
83
84 Args:
85 df: OHLCV DataFrame.
86 strategy: Strategy implementing the Strategy protocol.
87 symbol: Trading symbol.
88 initial_capital: Starting cash.
89 risk_per_trade: Fraction of equity risked per trade.
90
91 Returns:
92 PaperPortfolio with final state and trade log.
93 """
94 portfolio = PaperPortfolio(cash=initial_capital)
95 signals = strategy.generate_signals(df)
96
97 for i in range(1, len(signals)):
98 row = signals.iloc[i]
99 prev = signals.iloc[i - 1]
100
101 if row["Signal"] == 1 and prev["Signal"] != 1:
102 risk_amount = portfolio.equity * risk_per_trade
103 stop_distance = row["Close"] * 0.05
104 if stop_distance > 0:
105 shares = int(risk_amount / stop_distance)
106 if shares > 0:
107 portfolio.buy(symbol, float(row["Close"]), shares)
108
109 elif row["Signal"] == -1 and prev["Signal"] != -1:
110 for sym in list(portfolio.positions.keys()):
111 portfolio.sell(sym, float(row["Close"]))
112
113 final_price = float(signals["Close"].iloc[-1])
114 for sym in list(portfolio.positions.keys()):
115 portfolio.sell(sym, final_price)
116
117 return portfolio