When I correct for the collinearity of regressors in a multivariate
regression by replacing with residuals from another regression, regression in
Excel is not working. One or more of the coefficients in the multivariate
system (incorrectly) go to 0, with a p-value of 1. When I have tried on other
machines, it runs fine. It seems like my stat pack is flawed. Has anyone else
encountered this? Is there a way to debug?