一元模型拟合

 

import pandas as pd
import statsmodels.api as sm
# 数据的导入
data = pd.read_table("E:\\python\\part 2\\017\\TRD_Index.txt",sep='\t')

# 目标数据的提取

# 根据 行标签 提取数据
sh_data = data[data['Indexcd']==1]
sz_data = data[data.Indexcd==399106]
# 提取 列 数据
sh_ret = sh_data.Retindex
sz_ret = sz_data['Retindex']
# 索引一致化
# print(sh_ret.index)
# print(sz_ret.index)
sz_ret.index = sh_ret.index

# 构造回归模型
model = sm.OLS(sh_ret,sm.add_constant(sz_ret)).fit()
print(model.summary())

模型结果：

OLS Regression Results                            
==============================================================================
Dep. Variable:               Retindex   R-squared:                       0.825
Model:                            OLS   Adj. R-squared:                  0.825
Method:                 Least Squares   F-statistic:                     5698.
Date:                Thu, 14 Dec 2017   Prob (F-statistic):               0.00
Time:                        19:11:39   Log-Likelihood:                 4520.3
No. Observations:                1211   AIC:                            -9037.
Df Residuals:                    1209   BIC:                            -9026.
Df Model:                           1                                         
Covariance Type:            nonrobust                                         
==============================================================================
                 coef    std err          t      P>|t|      [95.0% Conf. Int.]
------------------------------------------------------------------------------
const         -0.0003      0.000     -1.747      0.081        -0.001  3.58e-05
Retindex       0.7603      0.010     75.487      0.000         0.741     0.780
==============================================================================
Omnibus:                      154.029   Durbin-Watson:                   1.821
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              382.039
Skew:                           0.701   Prob(JB):                     1.10e-83
Kurtosis:                       5.367   Cond. No.                         60.5
==============================================================================

Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Process finished with exit code 0

1. 可以得出R的平方为：0.825，可以解释82.5%的方差。
2. 截距项为-0.0003，p值为0.081，无法通过通过置信度为0.05的检验，可以推断该模型不含截距项，即为0.
3. 斜率为0.7603，通过检验，显著不为0.
4. 模型为：sh_ret = -0.0003 + 0.7603 sz_ret + e