Python量化交易学习笔记（十三）——按规则选股

下载数数据深沪6000多家公司、扩展因子、测试选股规则后，按照规则进行选股的过程。

选股规则

本文就选股规则所测试规则相同，即：

2日前倍量暴涨9%以上。
随后两日缩量调整。
收盘价在20日线上方。
20、30、60、120、250日线多头排列。
选股规则代码如下：

condition = df['value_boom_2a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] >= 2 * df['volume_3a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] > df['volume_1a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] > df['volume'].iloc[-1] and \
                    df['close'].iloc[-1] < df['close_2a'].iloc[-1] and \
                    df['close_1a'].iloc[-1] < df['close_2a'].iloc[-1] and \
                    df['close'].iloc[-1] > df['ma_20'].iloc[-1] and df['ma_30'].iloc[-1] > df['ma_60'].iloc[-1] \
                    > df['ma_120'].iloc[-1] > df['ma_250'].iloc[-1]

结果分析

选出的所有满足选股规则的股票，都会保存在csv文件中。
本文发布日期为2021年6月23日，当日只选出2只符合条件的股票，耗时880秒，csv文件截图如下。

?
1
2
3
4
C:\Python38\python.exe F:/test/src/com/gzh/demo49.py
sz002285
sz300627
程序所耗时间： 880.6239831447601

　　
　通了基于历史数据的规则验证以及选股流程。这样就可以先找到靠谱的规则，然后在每日盘后自动选出满足规则的股票，再做交易。有历史数据分析做支撑，至少不会成为韭菜里最高的那一波。

import datetime
import time
import sys
import os
import pandas as pd
 
# 获取当前目录
proj_path = os.path.dirname(os.path.abspath(sys.argv[0])) + '/../'
 
if __name__ == '__main__':
    # 程序开始时的时间
    time_start = time.time()
    # 读入股票代码
    stk_code_file = proj_path + 'data/tdx/all_codes.csv'
    codes = pd.read_csv(stk_code_file, encoding='unicode_escape')['code']
    code_list = []
    for code in codes:
        input_file = proj_path + 'data/extension/d/hard_rules/' + code + '.csv'
        df = pd.read_csv(input_file, index_col=0)
        condition = df['value_boom_2a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] >= 2 * df['volume_3a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] > df['volume_1a'].iloc[-1] and \
                    df['volume_2a'].iloc[-1] > df['volume'].iloc[-1] and \
                    df['close'].iloc[-1] < df['close_2a'].iloc[-1] and \
                    df['close_1a'].iloc[-1] < df['close_2a'].iloc[-1] and \
                    df['close'].iloc[-1] > df['ma_20'].iloc[-1] and df['ma_30'].iloc[-1] > df['ma_60'].iloc[-1] \
                    > df['ma_120'].iloc[-1] > df['ma_250'].iloc[-1]
        if condition:
            code_list.append(code)
            print(code)
    pd.DataFrame(data=code_list, columns=['code']).to_csv(proj_path + 'data/temp/' + datetime.datetime.strftime(
        datetime.datetime.now(), '%Y-%m-%d-%H-%M-%S') + '.csv', index=False)
    # 程序结束时系统时间
    time_end = time.time()
    print('程序所耗时间：', time_end - time_start)