红利低波策略

This is SUN Jiarui(Garrest)'s assignment for CUHKSZ MFE5210-Algorithm Trading lecture in 2024 Spring, i.e., Dividend and Low Volatility Strategy（Div'n'LowVolatStrategy）

通过对不同的条件进行筛选，筛选出具有高红利，低波动属性的股票池，构造投资组合，计算收益率

包安装准备

Some "pip install" preparation before running .py files, and my folder name in User is called "DELL". So on opening cmd, type command lines after C:\Users\DELL>

pip install --user iFinDAPI

THS_iFinDLogin('account','password')#my iFinD student account is mx_719562663

to complement a necessary package not locally included to support running, also: pip install --user concept_helper

基础数据池

DataFetcher()

DataFetcher() 类包含所有用于数据获取相关的函数，首先获取需要的数据

from stock_selector import DataFetcher
start_year = 2014
end_year = 2023
calender = DataFetcher().get_calender_data(start_year, end_year=2023, endlevel=[(1,2), (1), (1,2), (1,2)]) # 获取每月末交易日历
base_pool = pd.read("codes_map.csv") # 从数据文件夹中获取股票代码之间的映射表
mv_data = DataFetcher().get_mkv_data(start_year, end_year)
turnoverV_data = DataFetcher().get_turnoverV_data(start_year, end_year) # 已经是日均数据
dividend_index_data = DataFetcher().get_dividend_index_data

DataHandler

首先选定回测起止时间，基础股票池与每月月末交易日历，然后完成第一次筛选：

1. 过去一年日均总市值排名前80%
2. 过去一年日均成交额排名前80%
3. 连续三年分红

from stock_selector import DataHandler
mv_avg_data = DataHandler().mv_handler(mv_data, base_pool, calender) #获得240个交易日范围内日均市值数据
turnoverV_avg_data = DataHandler().turnoverV_handler(turnoverV_data, base_pool)
dividend_index = DataHandler().dividend_index_handler(dividend_index_data , base_pool, calender)

Selector()

base_pool = select_base_pool(mv_avg_data, turnoverV_avg_data, dividend_index)

筛选指标

DataFetcher()

股利支付率

$$ se = \frac{dps}{eps} = \frac{dps \times pe}{price} = pe \times dividendRatio $$

根据公式从JYDB.LC_DIndicesForValuation获取数据，计算每一期的股利支付率

股利增长率（3年）：

计算过去三年每股股利增长率，选取线性回归斜率大于0的股票

from stock_selector import DataFetcher()
payratio_data = DataFetcher().get_payratio_data(calender)
dividend_data = DataFetcher().get_dividend_data(start_year-3, end_year)

StockSelctor()

from stock_selector import DataHandler(), Selector()
payratio_data = DataHandler().payratio_handler(payratio_data, base_pool)
dividend_delta_data = DataHandler().dividend_delta_data(dividend_data, base_pool)
filter_pool_1 = Selector().filter_1(stocks_pool, payratio_data, devidend_delta_data)