Linear Regression | What is multivariate normality in assumptions of Linear Regression?
Question
TheDataMonk
Follow Me
About TheDataMonkGrand Master
I am the Co-Founder of The Data Monk. I have a total of 6+ years of analytics experience 3+ years at Mu Sigma 2 years at OYO 1 year and counting at The Data Monk I am an active trader and a logically sarcastic idiot :)Follow Me
Answers ( 2 )
Linear linear regression analysis requires all variables to be multivariate normal this can be best checked with the help of histogram for Q-Q plot(qqplot from statsmodel). I other words X and Y should be normally distributef.Normality can be checked using Kolmogrorov Smirnov test or simply by looking into the skewness and kurtosis for the given data. Also you can look into Shapiro Will test to know if the data was drawn from a Gaussian distribution sample. If the data is Not normally distributed you can use any non linear correction like log transform to make the data normally distributed
If the data points present on the plane which are normally disturbed along the regression line.If we plot histogram of residuals it will follow a normal distribution.