from xms_quant_backtrader import xms_quant_backtrader from tdx_strategy_models.small_fruit_band_trading import small_fruit_band_trading import pandas as pd from xg_tdx_func.xg_tdx_func import * class xms_quant_backtrader_zz100: def __init__(self, trader_tool='qmt', data_api='qmt', data_type='D', stock_list=[], start_date='20200101', end_date='20500101', total_cash=100000, st_name='全球外盘波段趋势增强策略回测' ): ''' 西蒙斯量化回测系统3.0 作者:西蒙斯量化 微信:xg_quant 数据源data_api选择qmt，需要先打开miniqmt登录，选择dfcf就不需要，建议使用qmt 股票池*************** 证券代码 名称 513,100 纳斯达克ETF 159,941 纳斯达克ETF广发 511,090 30年债券ETF 159,502 标普生物ETF 513,300 纳斯达克ETF 513,500 标普500ETF 159,937 黄金ETF 159,655 标普ETF 159,660 纳斯达克ETF 511,130 30年债券ETF 159,866 日本ETF 513,400 道琼斯ETF 511,800 黄金ETF 513,850 美国50ETF 511,260 10年债券ETF 159,659 纳斯达克ETF 513,290 纳斯达克生物ETF 159,501 纳斯达克ETF 159,934 黄金ETF 513,650 标普500ETF 511,520 政金债券ETF 513,520 日本ETF 513,730 东南亚ETF 159,687 亚太ETF 513,030 德国ETF 513,080 法国ETF ''' self.trader_tool=trader_tool self.data_api=data_api self.data_type=data_type self.stock_list=stock_list self.start_date=start_date self.end_date=end_date self.total_cash=total_cash self.trader=xms_quant_backtrader( trader_tool=self.trader_tool, data_api=self.data_api, data_type=self.data_type, stock_list=self.stock_list, start_date=self.start_date, end_date=self.end_date, total_cash=self.total_cash, st_name=st_name) self.hold_limit=5 self.buy_value=20000 self.sell_value=30000 def get_all_hist_data(self): ''' 获取全部历史数据 ''' df=pd.read_excel(r'外盘自定义交易股票池.xlsx') df['证券代码']=df['证券代码'].astype(str) stock_list=df['证券代码'].tolist() self.trader.re_trader_stock(stock_list=stock_list) self.trader.get_hist_data() hist=self.trader.data.get_hist_data() hist.to_csv(r'全部历史数据.csv') return hist def get_all_user_hist_data(self): ''' 加载本地历史数据 ''' stock_list=self.trader.get_read_tdx_data(r'ZZA100.blk')['证券代码'].tolist() self.trader.re_trader_stock(stock_list=stock_list) df=pd.read_csv(r'全部历史数据.csv') self.trader.get_user_hist_data(df) hist=self.trader.data.get_hist_data() return hist def cacal_all_stock_indicator(self): ''' 计算全部的指标 ''' data=pd.DataFrame() #hist=self.get_all_user_hist_data() hist=self.get_all_hist_data() print(hist) stock_list=self.trader.get_trader_stock() for stock in stock_list: df=hist[hist['stock']==stock] if df.shape[0]>0: models=small_fruit_band_trading(df=df) result=models.small_fruit_band_trading() stats=result['stats'].tolist() df['波段']=stats df.reset_index(drop=True, inplace=True) df['连续波段数量'] = BARSLASTCOUNT(df['波段']=='买') data=pd.concat([data,df],ignore_index=False) return data def get_buy_sell_data(self): ''' 获取买卖数据 ''' df=self.cacal_all_stock_indicator() buy_df=df[df['波段']=='买'] sell_df=df[df['波段']=='卖'] return buy_df,sell_df def run_backtrader(self): ''' 运行回测 ''' st_buy_df,st_sell_df=self.get_buy_sell_data() trader_date_list=self.trader.get_trader_date_list() for date in trader_date_list: if st_buy_df.shape[0]>0: buy_df=st_buy_df[st_buy_df['date']==date] #排序因子 buy_df=buy_df.sort_values(by='连续波段数量',ascending=False) else: buy_df=pd.DataFrame() if st_sell_df.shape[0]>0: sell_df=st_sell_df[st_sell_df['date']==date] else: sell_df=pd.DataFrame() position=self.trader.position.get_position_data() if position.shape[0]>0: position=position[position['股票余额']>=10] if position.shape[0]>0: position=position hold_amount=position.shape[0] hold_stock_list=position['证券代码'].tolist() else: position=pd.DataFrame() hold_amount=0 hold_stock_list=[] else: position=pd.DataFrame() hold_amount=0 hold_stock_list=[] if position.shape[0]>0: if sell_df.shape[0]>0: sell_stock_list=sell_df['stock'].tolist() else: sell_stock_list=[] position['卖出']=position['证券代码'].apply(lambda x: '是' if x in sell_stock_list else '不是') sell_df=position[position['卖出']=='是'] sell_amount=sell_df.shape[0] else: sell_df=pd.DataFrame() sell_amount=0 av_amount=(self.hold_limit-hold_amount)+sell_amount if av_amount<0: print(date,'达到持股限制不买入') av_amount=0 else: av_amount=av_amount print('***************************') print('{} 持股限制{} 持有数量{} 卖出数量{} 可以买入数量{}'.format(date,self.hold_limit,hold_amount,sell_amount,av_amount)) if buy_df.shape[0]>0: buy_df['持股']=buy_df['stock'].apply(lambda x: '是' if x in hold_stock_list else '不是') buy_df=buy_df[buy_df['持股']=='不是'] else: buy_df=buy_df buy_df=buy_df[:av_amount] print('持有的股票*****************') print(position) print('卖出股票***********') print(sell_df) print('买入股票************') print(buy_df) #先卖出在买入 if sell_df.shape[0]>0: for stock in sell_df['证券代码'].tolist(): price=self.trader.get_price(date=date,stock=stock) trader_type,amount,price=self.trader.order_value( date=date, stock=stock, value=self.sell_value, price=price, trader_type='sell' ) if trader_type=='sell' and amount>=10: self.trader.sell( date=date, stock=stock, amount=amount, price=price, maker='卖出成功') else: print(date,stock,'买入失败') else: print(date,'卖出没有持股数据') #买入 if buy_df.shape[0]>0: for stock in buy_df['stock'].tolist(): price=self.trader.get_open_price(date=date,stock=stock) trader_type,amount,price=self.trader.order_value( date=date, stock=stock, value=self.buy_value, price=price, trader_type='buy' ) if trader_type=='buy' and amount>=10: self.trader.buy( date=date, stock=stock, amount=amount, price=price, maker='买入成功') else: print(date,stock,'买入失败') else: print(date,'买入没有持股数据') #结算 self.trader.settlement_data(date=date) def get_backtrader_result(self): ''' 获取回测结果 ''' self.trader.generate_strategy_report() self.trader.get_all_trader_data() if __name__=='__main__': trader=xms_quant_backtrader_zz100() df=trader.run_backtrader() trader.get_backtrader_result()