temp.py

import os
# 必须在导入 akshare 之前禁用代理
os.environ['HTTP_PROXY'] = ''
os.environ['HTTPS_PROXY'] = ''
os.environ['http_proxy'] = ''
os.environ['https_proxy'] = ''

import akshare as ak
import pandas as pd
from dataclasses import dataclass, field
from functools import lru_cache
from requests.exceptions import RequestException
import time


def retry_request(func, max_retries=3, delay=2):
    """带重试的请求包装器"""
    for attempt in range(max_retries):
        try:
            return func()
        except RequestException as e:
            if attempt < max_retries - 1:
                print(f"网络请求失败，{delay}秒后重试 ({attempt + 1}/{max_retries})...")
                time.sleep(delay)
            else:
                raise e


@lru_cache(maxsize=1)
def get_stock_name_map() -> dict:
    """获取股票代码-名称映射表（带缓存）"""
    df = retry_request(lambda: ak.stock_info_a_code_name())
    return dict(zip(df['code'], df['name']))


def get_stock_name(symbol: str) -> str:
    """根据股票代码获取名称"""
    name_map = get_stock_name_map()
    return name_map.get(symbol, "未知")


@dataclass
class TradeConfig:
    """交易配置"""
    initial_capital: float = 100000  # 初始本金 10万
    buy_price: float = 3.8  # 低于此价格买入
    sell_price: float = 4.2  # 高于此价格卖出
    commission_rate: float = 0.00025  # 手续费万2.5
    min_commission: float = 5  # 最低手续费5元


@dataclass
class StockConfig:
    """股票配置"""
    symbol: str  # 股票代码
    start_date: str  # 开始日期 YYYYMMDD
    end_date: str  # 结束日期 YYYYMMDD
    name: str = field(default="")  # 股票名称（可选，留空自动获取）

    def __post_init__(self):
        if not self.name:
            self.name = get_stock_name(self.symbol)


def calc_commission(amount: float, config: TradeConfig) -> float:
    """计算交易手续费"""
    return max(amount * config.commission_rate, config.min_commission)


def simulate_trading(df: pd.DataFrame, config: TradeConfig) -> dict:
    """
    模拟交易
    策略：低于buy_price全仓买入，高于sell_price全部卖出
    """
    cash = config.initial_capital  # 现金
    shares = 0  # 持股数量
    buy_count = 0  # 买入次数
    sell_count = 0  # 卖出次数
    total_commission = 0  # 总手续费

    for _, row in df.iterrows():
        date = row['日期']
        close_price = row['收盘']

        if shares == 0 and close_price < config.buy_price:
            # 买入：全仓买入（按100股为一手）
            max_shares = int(cash / (close_price * 100)) * 100
            if max_shares > 0:
                cost = max_shares * close_price
                commission = calc_commission(cost, config)
                if cash >= cost + commission:
                    shares = max_shares
                    cash -= cost + commission
                    total_commission += commission
                    buy_count += 1
                    print(f"[{date}] 买入 {shares} 股, 价格 {close_price:.2f}, "
                          f"花费 {cost:.2f}, 手续费 {commission:.2f}")

        elif shares > 0 and close_price > config.sell_price:
            # 卖出：全部卖出
            revenue = shares * close_price
            commission = calc_commission(revenue, config)
            cash += revenue - commission
            total_commission += commission
            print(f"[{date}] 卖出 {shares} 股, 价格 {close_price:.2f}, "
                  f"收入 {revenue:.2f}, 手续费 {commission:.2f}")
            shares = 0
            sell_count += 1

    # 计算最终资产
    final_price = df.iloc[-1]['收盘']
    final_asset = cash + shares * final_price
    profit = final_asset - config.initial_capital
    profit_rate = (profit / config.initial_capital) * 100

    return {
        'final_asset': final_asset,
        'cash': cash,
        'shares': shares,
        'final_price': final_price,
        'profit': profit,
        'profit_rate': profit_rate,
        'buy_count': buy_count,
        'sell_count': sell_count,
        'total_commission': total_commission
    }


def run_backtest(stock_config: StockConfig, trade_config: TradeConfig):
    """运行单个股票的回测"""
    print("=" * 60)
    print(f"股票交易策略回测 - {stock_config.name}({stock_config.symbol})")
    print(f"策略: 低于 {trade_config.buy_price} 买入, 高于 {trade_config.sell_price} 卖出")
    print(f"本金: {trade_config.initial_capital:,.0f} 元")
    print(f"手续费: {trade_config.commission_rate*100:.4f}%, 最低 {trade_config.min_commission} 元")
    print("=" * 60)

    # 获取股票数据（带重试）
    print("\n正在获取股票数据...")
    try:
        df = retry_request(lambda: ak.stock_zh_a_hist(
            symbol=stock_config.symbol,
            period="daily",
            start_date=stock_config.start_date,
            end_date=stock_config.end_date,
            adjust=""  # 不复权
        ))
    except RequestException as e:
        print(f"获取股票数据失败: {e}")
        print("提示：如果遇到代理问题，请取消代码第11-12行的注释以禁用代理")
        return None

    print(f"获取到 {len(df)} 条交易日数据")
    print(f"日期范围: {df['日期'].iloc[0]} ~ {df['日期'].iloc[-1]}")
    print(f"价格范围: {df['收盘'].min():.2f} ~ {df['收盘'].max():.2f}")
    print()

    # 模拟交易
    print("交易记录:")
    print("-" * 60)
    result = simulate_trading(df, trade_config)

    # 输出结果
    print()
    print("=" * 60)
    print("回测结果")
    print("=" * 60)
    print(f"初始本金:     {trade_config.initial_capital:>15,.2f} 元")
    print(f"最终资产:     {result['final_asset']:>15,.2f} 元")
    print(f"  - 现金:     {result['cash']:>15,.2f} 元")
    print(f"  - 持股:     {result['shares']:>15} 股")
    print(f"  - 持股市值: {result['shares'] * result['final_price']:>15,.2f} 元")
    print(f"收益金额:     {result['profit']:>15,.2f} 元")
    print(f"收益率:       {result['profit_rate']:>14.2f}%")
    print(f"买入次数:     {result['buy_count']:>15} 次")
    print(f"卖出次数:     {result['sell_count']:>15} 次")
    print(f"总手续费:     {result['total_commission']:>15,.2f} 元")
    print("=" * 60)

    return result


def main():
    # ===== 股票配置（可添加多个股票进行测试）=====
    stocks = [
        StockConfig(symbol="000725", start_date="20250101", end_date="20251231"),
        # 添加更多股票（名称会自动获取）：
        # StockConfig(symbol="000001", start_date="20240101", end_date="20241231"),
        # StockConfig(symbol="000858", start_date="20240101", end_date="20241231"),
        # 或者手动指定名称：
        # StockConfig(symbol="600036", name="招商银行", start_date="20240101", end_date="20241231"),
    ]

    # ===== 交易配置 =====
    trade_config = TradeConfig(
        initial_capital=100000,  # 本金10万
        buy_price=3.8,          # 买入价格
        sell_price=4.2,         # 卖出价格
        commission_rate=0.00025, # 万2.5
        min_commission=5         # 最低5元
    )

    # 批量回测
    for stock in stocks:
        run_backtest(stock, trade_config)
        print("\n")


if __name__ == "__main__":
    main()