Wrong interpretation of the Shapiro-Wilk test
F-A opened this issue · 2 comments
F-A commented
In the Regression_diagnostics notebook , you are presenting the Shapiro-Wilk test.
The Shapiro-Wilk test's null hypothesis is that the data come from a Gaussian distribution. Therefore, the lower the p-value, the higher the change to reject the Gaussian distribution. The notebook says the opposite:
puddlejumper90 commented
Do you have a source article to reference? This might be good for those interested to study.
RRighart commented
Under "Encapsulation", cell 22 you have the line:
self.fitted_ = np.dot(X,mlr.coef_) + mlr.intercept_
Where I think "mlr" should be replaced by "self":
self.fitted_ = np.dot(X,self.coef_) + self.intercept_