HansDieter88 HansDieter88 - 3 months ago 16
R Question

how to calculate sum of squares separately for a regression

I´m trying to understand all the formulas around regression and am trying to calculate basically sum(xi-mean(x))^2 for a dataset.
Tried this:

sum(pubs$pubs-mean(pubs$pubs))^2

Answer

Usually you do the square first before summing, eg sum((y - yhat)^2); otherwise, you should get zero because mean(x) = sum(x)/n and sum(x-mean(x)) = sum(x) - n*mean(x).

Comments