I have been following SDS podcasts for around 8 months now but got to youtube channel just like a month ago. It is awesome seeing how quick you are getting new subscribers. Keep going :)
Then whats about correlation and Collinearity or multicollinearity ... If you take the square of that variable, it will cause correlation with X. which will make the regression biased. Should it be ok to use demean X in this case?
Cheers mate, it was an amazing explanation. Bytheway can you please tell me what software/ tool did you use to make this presentation. I mean data point Drawing
Hi, Polynomial regression of degree p in one independent variable x is considered. Numerically large sample correlations between xα and xβ, α < β, α, β = 1, ..., p, may cause ill- conditioning in the matrix to be inverted in application of the method of least squares. These sample correlations are investigated. It is confirmed that centering of the independent variable to have zero sample mean removes nonessential ill-conditioning. If the sample values of x are placed symmetrically about their mean, the sample correlation between xα and xβ is reduced to zero by centering when α + β is odd, but may remain large when α + β is even. Some examples and recommendations are given.