often said that the error term m in a regression equation represents the effect of the variables
that were omitted from the equation. This is unsatisfactory, even in simple contexts, as the following
discussion should indicate. Suppose subjects are IID, and all variables are jointly normal with
expectation 0. Suppose the explanatory variables have variance 1. The explanatory variables may
be correlated amongst themselves, but anyp of them have a non-singularp-dimensional distribution.
The parameters αj are real. Let