Python：如何评估 StatsModels 中的残差？
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/35417111/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Python: How to evaluate the residuals in StatsModels?
提问by DanielTheRocketMan
I want to evaluate the residuals: (y-hat y).
我想评估残差：（y-hat y）。
I know how to do that:
我知道该怎么做：
df = pd.read_csv('myFile', delim_whitespace = True, header = None)
df.columns = ['column1', 'column2']
y, X = ps.dmatrices('column1 ~ column2',data = df, return_type = 'dataframe')
model = sm.OLS(y,X)
results = model.fit()
predictedValues = results.predict()
#print predictedValues
yData = df.as_matrix(columns = ['column1'])
res = yData - predictedValues
I wonder if there is a Method to do this (?).
我想知道是否有一种方法可以做到这一点（？）。
采纳答案by TomAugspurger
That's stored in the residattribute of the Results class
这存储在Results 类的resid属性中
Likewise there's a results.fittedvaluesmethod, so you don't need the results.predict().
同样有一个results.fittedvalues方法，所以你不需要results.predict().
回答by SciPy
Normality of the residuals
残差的正态性
Option 1: Jarque-Bera test
选项 1：Jarque-Bera 测试
name = ['Jarque-Bera', 'Chi^2 two-tail prob.', 'Skew', 'Kurtosis']
test = sms.jarque_bera(results.resid)
lzip(name, test)
Out:
出去：
[('Jarque-Bera', 3.3936080248431666),
('Chi^2 two-tail prob.', 0.1832683123166337),
('Skew', -0.48658034311223375),
('Kurtosis', 3.003417757881633)]
Omni test:
Option 2: Omni test
选项 2：Omni 测试
name = ['Chi^2', 'Two-tail probability']
test = sms.omni_normtest(results.resid)
lzip(name, test)
Out:
出去：
[('Chi^2', 3.713437811597181), ('Two-tail probability', 0.15618424580304824)]
