Price difference analysis of the Python Arbitrage series--python Learning notes 22

# # #套利分析
Import Pandas as PD
Import NumPy as NP
#读取数据
Caipo=pd.read_excel (' caipo.xlsx ')
Doupo=pd.read_excel (' doupo.xlsx ')
Caipo.head (5)
#整理数据
caipo_close=caipo.ix[::,6]
Caipo_close.head ()
doupo_close=doupo.ix[::,6]
Doupo_close.head ()
"#分析数据
Cc=caipo_close
Dc=doupo_close
Cc.mean ()
Dc.mean ()
I=dc.mean ()/cc.mean ()
I
E=cc*i-dc
E.head ()
#基于均值比例分析残差

E.plot () ""

#发现基于均值比例的模型随机性太强, we recalculate based on the correlation coefficient.
Cc=caipo_close
Dc=doupo_close
Cc.corr (dc,method= ' Pearson ')
#从相关系数来看, the correlation between the two is very high, 94%;
#菜粕和豆粕价格相差很大, we calculate in different proportions
Ccnew=cc*123
dcnew=dc*100
E= (ccnew-dcnew)/100
E.head (10)

E.plot ()

Later on, we will make a KDJ-MACD model to calculate whether the price difference can be profitable.