Using the context of $(1)$ and $(2),$ please answer $(3).$

$(1)$ Fit a multiple linear regression model using all the rest of the $13$ variables as predictors $($after excluding variable $$subcontractor utilization$$ from the analysis$).$ Report all the variables with a p$-$value less than $0\mathrm{.}1$ in the format of $($variable name, p$-$value$).$

$(2)$ Run a forward selection procedure in SAS using all the rest of the $13$ variables as predictors $($after excluding variable $$subcontractor utilization$$ from the analysis$).$ Set $0\mathrm{.}1$ as the level of significance. Report the fitted regression model based on SAS output of Summary of Forward Selection.

$(3)$ The selected predictors from $(2)$ are quite different from those reported from $(1),$ even with the same cutoff $($alpha$=0\mathrm{.}1).$ Explain the discrepancy. Can you suggest a variable selection procedure that will lead to more similar results in terms of larger number of overlapping predictors? Explain why.