1. # coding=utf-8
      2. from __future__ import print_function, absolute_import, unicode_literals
      3. from gm.api import *
      5. '''
      6. 本策略每隔1个月定时触发计算SHSE.000300成份股的过去的EV/EBITDA并选取EV/EBITDA大于0的股票
      7. 随后平掉排名EV/EBITDA不在最小的30的股票持仓并等权购买EV/EBITDA最小排名在前30的股票
      8. 并用相应的CFFEX.IF对应的真实合约等额对冲
      9. 回测数据为:SHSE.000300和他们的成份股和CFFEX.IF对应的真实合约
      10. 回测时间为:2017-07-01 08:00:00到2017-10-01 16:00:00
      11. '''
      14. def init(context):
      15.     # 每月第一个交易日09:40:00的定时执行algo任务
      16.     schedule(schedule_func=algo, date_rule='1m', time_rule='09:40:00')
      18.     # 设置开仓在股票和期货的资金百分比(期货在后面自动进行杠杆相关的调整)
      19.     context.percentage_stock = 0.4
      20.     context.percentage_futures = 0.4
      23. def algo(context):
      24.     # 获取当前时刻
      25.     now = context.now
      26.     # 获取上一个交易日
      27.     last_day = get_previous_trading_date(exchange='SHSE', date=now)
      28.     # 获取沪深300成份股
      29.     stock300 = get_history_constituents(index='SHSE.000300', start_date=last_day,
      30.                                                 end_date=last_day)[0]['constituents'].keys()
      31.     # 获取上一个工作日的CFFEX.IF对应的合约
      32.     index_futures = get_continuous_contracts(csymbol='CFFEX.IF', start_date=last_day, end_date=last_day)[-1]['symbol']
      33.     # 获取当天有交易的股票
      34.     not_suspended_info = get_history_instruments(symbols=stock300, start_date=now, end_date=now)
      35.     not_suspended_symbols = [item['symbol'] for item in not_suspended_info if not item['is_suspended']]
      36.     # 获取成份股EV/EBITDA大于0并为最小的30个
      37.     fin = get_fundamentals(table='tq_sk_finindic', symbols=not_suspended_symbols,
      38.                            start_date=now, end_date=now, fields='EVEBITDA',
      39.                            filter='EVEBITDA>0', order_by='EVEBITDA', limit=30, df=True)
      40.     fin.index = fin.symbol
      41.     # 获取当前仓位
      42.     positions = context.account().positions()
      43.     # 平不在标的池或不为当前股指期货主力合约对应真实合约的标的
      44.     for position in positions:
      45.         symbol = position['symbol']
      46.         sec_type = get_instrumentinfos(symbols=symbol)[0]['sec_type']
      47.         # 若类型为期货且不在标的池则平仓
      48.         if sec_type == SEC_TYPE_FUTURE and symbol != index_futures:
      49.             order_target_percent(symbol=symbol, percent=0, order_type=OrderType_Market,
      50.                                  position_side=PositionSide_Short)
      51.             print('市价单平不在标的池的', symbol)
      52.         elif symbol not in fin.index:
      53.             order_target_percent(symbol=symbol, percent=0, order_type=OrderType_Market,
      54.                                  position_side=PositionSide_Long)
      55.             print('市价单平不在标的池的', symbol)
      57.     # 获取股票的权重
      58.     percent = context.percentage_stock / len(fin.index)
      59.     # 买在标的池中的股票
      60.     for symbol in fin.index:
      61.         order_target_percent(symbol=symbol, percent=percent, order_type=OrderType_Market,
      62.                              position_side=PositionSide_Long)
      63.         print(symbol, '以市价单调多仓到仓位', percent)
      65.     # 获取股指期货的保证金比率
      66.     ratio = get_history_instruments(symbols=index_futures, start_date=last_day, end_date=last_day)[0]['margin_ratio']
      67.     # 更新股指期货的权重
      68.     percent = context.percentage_futures * ratio
      69.     # 买入股指期货对冲
      70.     order_target_percent(symbol=index_futures, percent=percent, order_type=OrderType_Market,
      71.                          position_side=PositionSide_Short)
      72.     print(index_futures, '以市价单调空仓到仓位', percent)
      75. if __name__ == '__main__':
      76.     '''
      77.     strategy_id策略ID,由系统生成
      78.     filename文件名,请与本文件名保持一致
      79.     mode实时模式:MODE_LIVE回测模式:MODE_BACKTEST
      80.     token绑定计算机的ID,可在系统设置-密钥管理中生成
      81.     backtest_start_time回测开始时间
      82.     backtest_end_time回测结束时间
      83.     backtest_adjust股票复权方式不复权:ADJUST_NONE前复权:ADJUST_PREV后复权:ADJUST_POST
      84.     backtest_initial_cash回测初始资金
      85.     backtest_commission_ratio回测佣金比例
      86.     backtest_slippage_ratio回测滑点比例
      87.     '''
      88.     run(strategy_id='strategy_id',
      89.         filename='main.py',
      90.         mode=MODE_BACKTEST,
      91.         token='token_id',
      92.         backtest_start_time='2017-07-01 08:00:00',
      93.         backtest_end_time='2017-10-01 16:00:00',
      94.         backtest_adjust=ADJUST_PREV,
      95.         backtest_initial_cash=10000000,
      96.         backtest_commission_ratio=0.0001,
      97.         backtest_slippage_ratio=0.0001)

    alpha对冲(股票+期货)策略效果

    原文: https://www.myquant.cn/docs/python_strategyies/101