改用marimo编辑notebook方便跨平台使用

notebook/backtest.py

@@ -0,0 +1,184 @@
+import marimo
+
+__generated_with = "0.19.6"
+app = marimo.App(width="full", auto_download=["ipynb"], sql_output="pandas")
+
+
+@app.cell
+def _():
+    import marimo as mo
+    return (mo,)
+
+
+@app.cell
+def _():
+    import urllib
+    import sqlalchemy
+
+    host = "81.71.3.24"
+    port = 6785
+    username = "leopard"
+    password = urllib.parse.quote_plus("9NEzFzovnddf@PyEP?e*AYAWnCyd7UhYwQK$pJf>7?ccFiN^x4$eKEZ5~E<7<+~X")
+    database = "leopard_dev"
+    engine = sqlalchemy.create_engine(f"postgresql://{username}:{password}@{host}:{port}/{database}")
+    return (engine,)
+
+
+@app.cell
+def _(engine, mo):
+    dailies_df = mo.sql(
+        f"""
+        select trade_date,
+               open * factor         as Open,
+               close * factor        as Close,
+               high * factor         as High,
+               low * factor          as Low,
+               volume                as Volume,
+               coalesce(factor, 1.0) as factor
+        from leopard_daily daily
+                 left join leopard_stock stock on stock.id = daily.stock_id
+        where stock.code = '000001.SZ'
+          and daily.trade_date between '2024-01-01 00:00:00' and '2025-12-31 23:59:59'
+        order by daily.trade_date
+        """,
+        engine=engine
+    )
+    return (dailies_df,)
+
+
+@app.cell
+def _(dailies_df):
+    import pandas as pd
+
+    dailies_df.rename(
+        columns={'open': 'Open', 'close': 'Close', 'high': 'High', 'low': 'Low', 'volume': 'Volume'}, inplace=True
+    )
+    dailies_df['trade_date'] = pd.to_datetime(dailies_df['trade_date'], format='%Y-%m-%d')
+    dailies_df.set_index('trade_date', inplace=True)
+    dailies_df
+    return
+
+
+@app.cell
+def _(dailies_df):
+    import talib
+
+    dailies_df['sma10'] = talib.SMA(dailies_df['Close'], timeperiod=10)
+    dailies_df['sma30'] = talib.SMA(dailies_df['Close'], timeperiod=30)
+    dailies_df['sma60'] = talib.SMA(dailies_df['Close'], timeperiod=60)
+    dailies_df['sma120'] = talib.SMA(dailies_df['Close'], timeperiod=120)
+    dailies_df
+    return
+
+
+@app.cell
+def _(dailies_df):
+    # 指标计算完成后截取后面指标的完整的部份使用
+    target_dailies_df = dailies_df.loc['2025-01-01':'2025-12-31']
+    target_dailies_df
+    return (target_dailies_df,)
+
+
+@app.cell
+def _():
+    from backtesting import Strategy
+    from backtesting.lib import crossover
+
+    class SmaCross(Strategy):
+        def init(self):
+            self.sma10 = self.I(lambda x: x, self.data.sma10)
+            self.sma30 = self.I(lambda x: x, self.data.sma30)
+            self.sma60 = self.I(lambda x: x, self.data.sma60)
+            # self.sma120 = self.I(lambda x: x, self.data.sma120)
+
+        def next(self):
+            if self.sma60 > 0 and crossover(self.data.sma10, self.data.sma30):
+                self.buy()
+            elif self.position.size > 0 and crossover(self.data.sma30, self.data.sma10):
+                self.position.close()
+    return (SmaCross,)
+
+
+@app.function
+def stats_print(stats):
+    indicator_name_mapping = {
+        # 'Start': '回测开始时间',
+        # 'End': '回测结束时间',
+        # 'Duration': '回测持续时长',
+        # 'Exposure Time [%]': '持仓时间占比（%）',
+        'Equity Final [$]': '最终收益',
+        'Equity Peak [$]': '峰值收益',
+        'Return [%]': '总收益率（%）',
+        'Buy & Hold Return [%]': '买入并持有收益率（%）',
+        'Return (Ann.) [%]': '年化收益率（%）',
+        'Volatility (Ann.) [%]': '年化波动率（%）',
+        # 'CAGR [%]': '复合年均增长率（%）',
+        # 'Sharpe Ratio': '夏普比率',
+        'Sortino Ratio': '索提诺比率',
+        'Calmar Ratio': '卡尔玛比率',
+        # 'Alpha [%]': '阿尔法系数（%）',
+        # 'Beta': '贝塔系数',
+        'Max. Drawdown [%]': '最大回撤（%）',
+        'Avg. Drawdown [%]': '平均回撤（%）',
+        'Max. Drawdown Duration': '最大回撤持续时长',
+        'Avg. Drawdown Duration': '平均回撤持续时长',
+        '# Trades': '总交易次数',
+        'Win Rate [%]': '胜率（%）',
+        # 'Best Trade [%]': '最佳单笔交易收益率（%）',
+        # 'Worst Trade [%]': '最差单笔交易收益率（%）',
+        # 'Avg. Trade [%]': '平均单笔交易收益率（%）',
+        # 'Max. Trade Duration': '单笔交易最长持有时长',
+        # 'Avg. Trade Duration': '单笔交易平均持有时长',
+        # 'Profit Factor': '盈利因子',
+        # 'Expectancy [%]': '期望收益（%）',
+        'SQN': '系统质量数',
+        # 'Kelly Criterion': '凯利准则',
+    }
+    for k, v in stats.items():
+        if k in indicator_name_mapping:
+            cn_name = indicator_name_mapping.get(k, k)
+            if isinstance(v, (int, float)):
+                if "%" in cn_name or k in ['Sharpe Ratio', 'Sortino Ratio', 'Calmar Ratio', 'Profit Factor']:
+                    formatted_value = f"{v:.2f}"
+                elif "$" in cn_name:
+                    formatted_value = f"{v:.2f}"
+                elif "次数" in cn_name:
+                    formatted_value = f"{v:.0f}"
+                else:
+                    formatted_value = f"{v:.4f}"
+            else:
+                formatted_value = str(v)
+            print(f'{cn_name}: {formatted_value}')
+
+
+@app.cell
+def _(SmaCross, target_dailies_df):
+    from backtesting import Backtest
+
+    import backtesting._plotting as plotting
+    from bokeh.colors.named import tomato, lime
+
+    plotting.BULL_COLOR = tomato
+    plotting.BEAR_COLOR = lime
+
+    bt = Backtest(target_dailies_df, SmaCross, cash=100000, commission=.002, finalize_trades=True)
+    stats = bt.run()
+    stats_print(stats)
+    return bt, stats
+
+
+@app.cell
+def _(stats):
+    stats._trades
+    # stats._equity_curve
+    return
+
+
+@app.cell
+def _(bt):
+    bt.plot()
+    return
+
+
+if __name__ == "__main__":
+    app.run()