DataFrame 添加索引并利用协方差计算

F数据格式：

beta,1.7123128642991907
industry1,0.23997950482951863
industry2,-0.23399748148492977
industry3,1.0358843834562912
industry4,0.2558507224304266
industry5,1.0172139863299385
industry6,1.808718895207218
industry7,0.9461830555049734
industry8,0.7578003666560819
industry9,0.7849956139973872
industry10,-0.4826768145696424
industry11,0.13532290488995724
industry12,0.49963746964732053
industry13,1.5749337193085124
industry14,0.1860840097236171
industry15,0.30044258386704364
industry16,1.197740076784449
industry17,0.10193938507719173
industry18,0.46973081299889374
industry19,0.31897646614614783
industry20,1.872140058959709
industry21,0.6118553140854857
industry22,0.7457968747058069
industry23,0.028222110889729896
industry24,-0.22644238223979166
industry25,0.542633477760374
industry26,-1.401701678495885
industry27,1.6684175316424383
industry28,1.393635068564524
market_value,-0.10610649438327055
ep,-0.00034028733231844174

import numpy as np
import pandas as pd
f1 = [1,2,3,4]
f2 = [2,4,8,9]
corr = np.corrcoef(f1,f2)
print('corr',corr)





def get_corr(df_x):
    return df_x.corr()

arr = np.array([f1,f2])


# print('f1.cols',f1)
f_data01 = pd.read_csv("../F_2016-03-01.csv", names=["factor", "date01"]).set_index("factor")
f_data02 = pd.read_csv("../F_2016-03-02.csv", names=["factor", "date02"]).set_index("factor")
f_data03 = pd.read_csv("../F_2016-03-03.csv", names=["factor", "date03"]).set_index("factor")
f_data04 = pd.read_csv("../F_2016-03-04.csv", names=["factor", "date04"]).set_index("factor")
# print(f_data01)
f_data = pd.concat([f_data01, f_data02, f_data03, f_data04], axis=1)

# print('df_x',f_data)
df_x_T = f_data.T
#print('df_x_T',df_x_T)
t1 = df_x_T.corr()
print ('df_x',t1)