Python 股票数据分析 tushare获取数据 统计散点图折线图 线性回归建模 三因子模型R market SMB HML

import numpy as np import pandas as pd import matplotlib from matplotlib import pyplot as plt font = { "family": "Microsoft YaHei" } matplotlib.rc("font", **font) import statsmodels.api as sm # 读取数据 shangzheng = pd.read_csv('shangzheng.csv') # 计算retindex shangzheng['Retindex'] = np.log(shangzheng['收盘价'] / shangzheng['收盘价'].shift(1))[1:] xinan = pd.read_csv('xinan.csv') xinan['Retindex'] = np.log(xinan['close'] / xinan['close'].shift(1))[1:] # 转换数据类型为string xinan['trade_date'] = xinan['trade_date'].apply(lambda x: str(x)) # 把2014/1/1格式的日期转换为20140101格式 def date(x): ymd = x.split('/') year = ymd[0] month = ymd[1] if len(ymd[1]) == 2 else '0' + ymd[1] day = ymd[2] if len(ymd[2]) == 2 else '0' + ymd[2] return year + month + day shangzheng['日期'] = shangzheng['日期'].apply(date) # 转换完以后可以用相同日期为基准合并两张表 shangzheng_xinan = pd.merge(shangzheng, xinan, left_on='日期', right_on='trade_date') # 取出需要的3列 ret = shangzheng_xinan.loc[:, ['日期', 'Retindex_x', 'Retindex_y']] # 去除异常空值 ret = ret.dropna() # 计算eret rf = 1.036 ** (1 / 360) - 1 ret['eret_x'] = ret['Retindex_x'].apply(lambda x: x - rf) ret['eret_y'] = ret['Retindex_y'].apply(lambda x: x - rf) # 画散点图 plt.scatter(ret['eret_x'], ret['eret_y'], color='black') plt.xlim(-0.06, 0.06) plt.ylim(-0.15, 0.15) plt.title('XinAnGuFen return and market return') plt.savefig('XinAnGuFen return and market return.png') plt.show() # 计算OLS model = sm.OLS(ret['eret_y'], sm.add_constant(ret['eret_x'])) result = model.fit() print(result.summary()) # 读数据，计算retindex huaxia = pd.read_csv('huaxia.csv') ret['Retindex_huaxia'] = np.log(huaxia['close'] / huaxia['close'].shift(1))[1:] # 画折线图 retindex plt.plot(ret['Retindex_huaxia'], color='black') plt.savefig('huaxia2014ret.png') plt.show() # 日期转为string huaxia['trade_date'] = huaxia['trade_date'].apply(lambda x: str(x)) # 组合华夏银行和前面的ret表 huaxia_ret = pd.merge(huaxia, ret, left_on='trade_date', right_on='日期') # 读三因子数据 yinzi = pd.read_csv('三因子example/chuangyeban3yinzi.csv').iloc[1:, 2:] yinzi = yinzi.reset_index() # 拼接ret和三因子两张表 hxthrfac = pd.concat([huaxia_ret['Retindex_huaxia'], yinzi.iloc[:, 1:]], axis=1) # 画散点图 plt.scatter(hxthrfac['Retindex_huaxia'], hxthrfac['R_market']) plt.savefig('R_market.png') plt.show() plt.scatter(hxthrfac['Retindex_huaxia'], hxthrfac['SMB']) plt.savefig('SMB.png') plt.show() plt.scatter(hxthrfac['Retindex_huaxia'], hxthrfac['HML']) plt.savefig('HML.png') plt.show() # 计算OLS regthrfac = sm.OLS(hxthrfac['Retindex_huaxia'], sm.add_constant(hxthrfac.iloc[:, 1:])) result = regthrfac.fit() print(result.summary())