excellent video. i got a question. i would like to use this method to test if my regression model's residuals are normally distributed or not. since the classic assumption of a regression model is having a normally distributed residuals N(0,sigma^2). the question is, should i test shapiro wilk onto the residuals of the model or the standardized residuals? to proof that our regression model is indeed have residuals normally distributed. for the significance level are 5%, my residuals are not normally distributed, having p-value of 0,039999. however for the standardized residuals i got p-value of 0,06807, meaning that my standardized residuals are indeed normally distributed
Hi, L. Thanks for watching my youtube contents. For your 1st question, the Shapiro Wilk test is useful for normality test. So, you can use this test for raw data or residual data from regression model. For your 2nd question, I would say you need to use this test for residual because the main purpose of analysing standardized residuals is to detect outliers. But, I couldn't find proper site or article to support my comments. Please let me know if you find useful paper or link. Many thanks, Sangwoo.
@@sangwoo.statistics thank you for the answer. i think about that aswell.. another question or should i say a request maybe haha. could you perhaps do a method on finding p value of durbin watson test using excel ? that would greatly helps
excellent video. i got a question. i would like to use this method to test if my regression model's residuals are normally distributed or not. since the classic assumption of a regression model is having a normally distributed residuals N(0,sigma^2).
the question is, should i test shapiro wilk onto the residuals of the model or the standardized residuals? to proof that our regression model is indeed have residuals normally distributed.
for the significance level are 5%, my residuals are not normally distributed, having p-value of 0,039999. however for the standardized residuals i got p-value of 0,06807, meaning that my standardized residuals are indeed normally distributed
Hi, L. Thanks for watching my youtube contents. For your 1st question, the Shapiro Wilk test is useful for normality test. So, you can use this test for raw data or residual data from regression model. For your 2nd question, I would say you need to use this test for residual because the main purpose of analysing standardized residuals is to detect outliers. But, I couldn't find proper site or article to support my comments. Please let me know if you find useful paper or link. Many thanks, Sangwoo.
@@sangwoo.statistics thank you for the answer. i think about that aswell..
another question or should i say a request maybe haha. could you perhaps do a method on finding p value of durbin watson test using excel ? that would greatly helps
very useful .. thank you!
Excellent video...