Python——量化分析介绍（十一）

├── MyQuant_v1 #量化分析程序目录
   ├── __init__.py
   ├── data #数据处理目录
   │   ├── __init__.py

   │   ├── basic_crawler.py# 爬取股票基础信息存入MongoDB数据库.

   │   ├── data_crawler.py  #爬取指数、股票数据

   │   ├── data_module.py  #从数据集中获取股票数据

   │   └── finance_report_crawler.py  #爬取财报数据

   ├──util # 公用程序
   │   ├── __init__.py

   │   ├── stock_util.py#获取股票交易日期、前一交易日日期、股票代码

   │   └── database.py #链接数据库

   ├── factor #因子
   │   ├── __init__.py

   │   └── pe_factor.py#计算各股票的止盈率，用于后面的股票池策略

   ├── strategy #策略
   │   ├── __init__.py

   └─── stock_pool_strategy.py #股票池策略及效果演示

讲一下最后这节的干货吧：stock_pool_strategy.py #股票池策略及效果演示

#!/usr/bin/env python3.6
# -*- coding: utf-8 -*-
# @Time    : 2019-08-05 21:47
# @Author  : Ed Frey
# @File    : stock_pool_strategy.py
# @Software: PyCharm
"""
实现股票池，条件是0 < PE <30， 按照PE正序排列，最多取100只票；
再平衡周期为7个交易日
主要的方法包括：
stock_pool：找到两个日期之间所有出的票
find_out_stocks：找到当前被调出的股票
evaluate_stock_pool：对股票池的性能做初步验证，确保股票池能够带来Alpha
"""
from pymongo import ASCENDING
import pandas as pd
import matplotlib.pyplot as plt
from util.database import DB_CONN
from util.stock_util import get_trading_dates

daily = DB_CONN['daily']
daily_hfq = DB_CONN['daily_hfq']

def stock_pool(begin_date, end_date):
    """
    股票池
    :param begin_date: 开始日期
    :param end_date: 结束日期
    :return: tuple，所有调整日，以及调整日和代码列表对应的dict
    """
    adjust_date_codes_dict = dict()
    all_dates =  get_trading_dates(begin_date=begin_date, end_date=end_date)
    last_phase_codes = []
    adjust_interval = 7
    all_adjust_dates = []

    for _index in range(0, len(all_dates), adjust_interval):
        adjust_date = all_dates[_index]
        all_adjust_dates.append(adjust_date)
        # print('调整日期：%s' % adjust_date, flush=True)
        daily_cursor = daily.find(
            {'date': adjust_date, 'pe': {'$lt': 30, '$gt': 0},
             'index':False, 'is_trading': True},
            sort=[('pe', ASCENDING)],
            projection={'code': True},
            limit=100
        )

        codes = [x['code'] for x in daily_cursor]
        this_phase_codes = []
        if len(last_phase_codes) > 0:
            suspension_cursor = daily.find(
                {'code': {'$in': last_phase_codes}, 'date': adjust_date, 'is_trading': False},
                projection={'code': True}
            )
            suspension_codes = [x['code'] for x in suspension_cursor]
            this_phase_codes = suspension_codes
        # print('上期停牌', flush=True)
        # print(this_phase_codes, flush=True)
        this_phase_codes += codes[0: 100 - len(this_phase_codes)]
        last_phase_codes = this_phase_codes
        adjust_date_codes_dict[adjust_date] = this_phase_codes
        # print('最终出票', flush=True)
        # print(this_phase_codes, flush=True)
    return all_adjust_dates, adjust_date_codes_dict


def evaluate_stock_pool():
    """
    对股票池做一个简单的评价
    """
    # 设定评测周期
    adjust_dates, codes_dict = stock_pool('2018-01-01', '2019-07-28')
    # 用DataFrame保存收益
    df_profit = pd.DataFrame(columns=['profit', 'hs300'])
    df_profit.loc[adjust_dates[0]] = {'profit': 0, 'hs300': 0}
    hs300_begin_value = daily.find_one({'code': '000300', 'index': True, 'date': adjust_dates[0]})['close']

    # 通过净值计算累计收益
    net_value = 1
    for _index in range(1, len(adjust_dates) - 1):
        last_adjust_date = adjust_dates[_index - 1]
        current_adjust_date = adjust_dates[_index]
        # 获取上一期的股票池
        codes = codes_dict[last_adjust_date]

        # 构建股票代码和后复权买入价格的股票
        code_buy_close_dict = dict()
        buy_daily_cursor = daily_hfq.find(
            {'code': {'$in': codes}, 'date': last_adjust_date,'index':False},
            projection={'close': True, 'code': True}
        )

        for buy_daily in buy_daily_cursor:
            code = buy_daily['code']
            code_buy_close_dict[code] = buy_daily['close']

        # 获取到期的股价
        sell_daily_cursor = daily_hfq.find(
            {'code': {'$in': codes}, 'date': current_adjust_date,'index':False},
            projection={'close': True, 'code': True}
        )

        # 计算单期收益
        profit_sum = 0
        count = 0
        for sell_daily in sell_daily_cursor:
            code = sell_daily['code']

            if code in code_buy_close_dict:
                buy_close = code_buy_close_dict[code]
                sell_close = sell_daily['close']

                profit_sum += (sell_close - buy_close) / buy_close

                count += 1
        print("收益报告：%s，%s"%(current_adjust_date,profit_sum))
        if count > 0:
            profit = round(profit_sum / count, 4)
            hs300_close = daily.find_one({'code': '000300', 'index': True, 'date': current_adjust_date})['close']

            net_value = net_value * (1 + profit)
            df_profit.loc[current_adjust_date] = {
                'profit': round((net_value - 1) * 100, 4),
                'hs300': round((hs300_close - hs300_begin_value) * 100 / hs300_begin_value, 4)}
            print(df_profit)

    df_profit.plot(title='Stock Pool Evaluation Result', kind='line')
    df_profit.plot()
    plt.show()


if __name__ == "__main__":

    # stock_pool('2018-01-01', '2019-07-28')
    evaluate_stock_pool()