量化之趋势跟踪策略

客官°小女子只卖身不卖艺 2021-07-27 02:57 523阅读 0赞

趋势跟踪策略，目前逻辑太简单，效果不怎么样。

一种是手写

import numpy as np
    import pandas as pd
    import matplotlib as mpl
    import matplotlib.pyplot as plt
    import seaborn as sns
    from finance import GetData
    import warnings;
    warnings.filterwarnings('ignore')
    
    data = GetData.get_data('../data/002236.csv')
    data_2 = GetData.get_data('../data/002415.csv')
    # 选定使用特斯拉两年的股票走势数据
    # 头一年（[:252]）作为训练数据, 美股交易中一年的交易日有252天
    train_kl = data[-900:]
    # 后一年（[252:]）作为回测数据
    test_kl = data[-400:-200]
    buy_signal=0
    sell_signal=0
    kl_pd=train_kl
    def shape():
        sns.set_context(rc={'figure.figsize': (14, 7)})
        sns.regplot(x=np.arange(0, data.shape[0]), y=data.close.values, marker='+')
        plt.show()
    def train_plot():
        start = 0
        # 前1/4的数据
        end = int(kl_pd.shape[0] / 4)
        # 将x也使用arange切割
        x = np.arange(start, end)
        # y根据start，end进行切片
        y = kl_pd.close.values[start:end]
        sns.regplot(x=x, y=y, marker='+')
        # plt.show()
    
        # 当天收盘价格超过N1天内最高价格作为买入信号
        N1 = 42
        # 当天收盘价格超过N2天内最低价格作为卖出信号
        N2 = 41
    
        from abupy import pd_rolling_max
        # rolling_max示例序列
        demo_list = np.array([1, 2, 1, 1, 100, 1000])
        # 对示例序列以3个为一组，寻找每一组中的最大值
        # pd.rolling_max(demo_list, window=3)
        pd_rolling_max(demo_list, window=3)
    
        # expanding_max示例序列
        demo_list = np.array([1, 2, 1, 1, 100, 1000])
        from abupy import pd_expanding_max
        # pd.expanding_max(demo_list)
        pd_expanding_max(demo_list)
    
        # 通过rolling_max方法计算最近N1个交易日的最高价
        # kl_pd['n1_high'] = pd.rolling_max(kl_pd['high'], window=N1)
        kl_pd['n1_high'] = pd_rolling_max(kl_pd['high'], window=N1)
        # 表7-4所示
        print(kl_pd[0:5])
    
        # expanding_max
        # expan_max = pd.expanding_max(kl_pd['close'])
        expan_max = pd_expanding_max(kl_pd['close'])
        # fillna使用序列对应的expan_max
        kl_pd['n1_high'].fillna(value=expan_max, inplace=True)
        # 表7-5所示
        kl_pd[0:5]
    
        from abupy import pd_rolling_min, pd_expanding_min
        # 通过rolling_min方法计算最近N2个交易日的最低价格
        # rolling_min与rolling_max类似
        # kl_pd['n2_low'] = pd.rolling_min(kl_pd['low'], window=N2)
        kl_pd['n2_low'] = pd_rolling_min(kl_pd['low'], window=N2)
        # expanding_min与expanding_max类似
        # expan_min = pd.expanding_min(kl_pd['close'])
        expan_min = pd_expanding_min(kl_pd['close'])
        # fillna使用序列对应的eexpan_min
        kl_pd['n2_low'].fillna(value=expan_min, inplace=True)
    
        # 当天收盘价格超过N天内的最高价或最低价, 超过最高价格作为买入信号买入股票持有
        buy_index = kl_pd[kl_pd['close'] > kl_pd['n1_high'].shift(1)].index
        kl_pd.loc[buy_index, 'signal'] = 1
    
        # 当天收盘价格超过N天内的最高价或最低价, 超过最低价格作为卖出信号
        sell_index = kl_pd[kl_pd['close'] < kl_pd['n2_low'].shift(1)].index
        kl_pd.loc[sell_index, 'signal'] = 0
    
        kl_pd.signal.value_counts().plot(kind='pie', figsize=(5, 5));
        # plt.show()
    
        """
            将信号操作序列移动一个单位，代表第二天再将操作信号执行，转换得到持股状态
            这里不shift(1)也可以，代表信号产生当天执行，但是由于收盘价格是在收盘后
            才确定的，计算突破使用了收盘价格，所以使用shift(1)更接近真实情况
        """
        kl_pd['keep'] = kl_pd['signal'].shift(1)
        kl_pd['keep'].fillna(method='ffill', inplace=True)
    
        # 计算基准收益
        kl_pd['benchmark_profit'] = np.log(
            kl_pd['close'] / kl_pd['close'].shift(1))
    
        # 计算使用趋势突破策略的收益
        kl_pd['trend_profit'] = kl_pd['keep'] * kl_pd['benchmark_profit']
    
        # 可视化收益的情况对比
        kl_pd[['benchmark_profit', 'trend_profit']].cumsum().plot(grid=True,
                                                                  figsize=(
                                                                      14, 7));
        plt.show()

一种是依靠掘金客户端

# coding=utf-8
    from __future__ import print_function, absolute_import, unicode_literals
    
    from gm.api import *
    
    
    '''
    策略思路：
    前一天收盘价格超过N1天内最高价格作为买入信号
    前一天收盘价格超过N2天内最低价格作为卖出信号
    '''
    
    
    def init(context):
        # 每月第一个交易日的09:40 定时执行algo任务
        schedule(schedule_func=algo, date_rule='1d', time_rule='09:40:00')
        # 数据滑窗
        # context.date = 5
        # 设置开仓的最大资金量
        context.ratio = 0.8
        # 沪深300 SHSE.000300，创业板指 SZSE.399006
        context.symbol = "SHSE.603259"
        context.num = 5
        context.list =[]
        # 当天收盘价格超过N1天内最高价格作为买入信号
        context.n1 = 22
        # 当天收盘价格超过N2天内最低价格作为卖出信号
        context.n2 = 11
        # 订阅行情 第一个参数是标的， 第二个是时间周期，表示日线，第三个指指定设置count参数，表示需要的滑窗大
        # 详见 https://www.myquant.cn/docs/python/python_subscribe#15ad56f8be8519c0
        subscribe(symbols=context.symbol, frequency='1d', count=context.n1)
    
    def algo(context):
        now = context.now
        symbol = context.symbol
        # 获取数据滑窗，只要在init里面有订阅，在这里就可以取的到， 返回值是pandas.DataFrame的数据结构
        # data_n1 = context.data(symbol=symbol, frequency='1d', count=context.n1, fields='high')
        data_n1 = history_n(symbol=symbol, frequency='1d', end_time=now, fields='high,low,close', count=context.n1, df=True)
        high = max(data_n1['high'].values[:-1])
        high=round(high,2)
    
        # data_n2 = context.data(symbol=symbol, frequency='1d', count=context.n2, fields='low,close')
        low = max(data_n1['low'].values[:-context.n2])
        low = round(low, 2)
    
        price = data_n1['close'].values[-1]
        price = round(price, 2)
        print(now,'\n high:',high,' low:',low,' price:',price)
        position_long = context.account().position(symbol=symbol, side=PositionSide_Long)
    
        if (not  position_long and price > high ):
            context.list.append(symbol)
            # order_target_percent(symbol=symbol, percent=1/context.num, order_type=OrderType_Market,
            #                      position_side=PositionSide_Long)
            # print(str(now) + '买入：\n' + str(symbol))
        if (position_long and price > low):
            order_target_percent(symbol=symbol, percent=0, order_type=OrderType_Market,
                                 position_side=PositionSide_Long)
            print(str(now) + '卖出：\n' + str(symbol))
    # 查看最终的回测结果
    def on_backtest_finished(context, indicator):
        print(indicator)
    
    if __name__ == '__main__':
        '''
        strategy_id策略ID,由系统生成
        filename文件名,请与本文件名保持一致
        mode实时模式:MODE_LIVE回测模式:MODE_BACKTEST
        token绑定计算机的ID,可在系统设置-密钥管理中生成
        backtest_start_time回测开始时间
        backtest_end_time回测结束时间
        backtest_adjust股票复权方式不复权:ADJUST_NONE前复权:ADJUST_PREV后复权:ADJUST_POST
        backtest_initial_cash回测初始资金
        backtest_commission_ratio回测佣金比例
        backtest_slippage_ratio回测滑点比例
        '''
    
        run(strategy_id='35693ff8-e873-11ea-aaa0-00ff0e1a51ea',
            filename='trend_following.py',
            mode=MODE_BACKTEST,
            token='9b06ce1b5a39ac39d71e58836549a99fdcb60',
            backtest_start_time='2019-01-01 08:00:00',
            backtest_end_time='2020-08-27 16:00:00',
            backtest_adjust=ADJUST_PREV,
            backtest_initial_cash=1000000,
            backtest_commission_ratio=0.0003,
            backtest_slippage_ratio=0.0001)