Regression assumptions

noetsi

Fortran must die
#1
This is a strange question I know. I work for a state agency that in practice works for a federal one. They have a model that is going to be run even if there are problems with the assumptions in the model. That is the federal government is going to run the model whether there are errors in the assumptions or not. It is possible that they will analyze the residuals and correct problems. But I don't know they will and they have never brought this up with us.

So if we find these problems in our model, and they are going to run it anyhow, does it matter? :p I guess for our own auxillary analysis we should still correct errors even if they do not. Formally we have our entire population, one could argue this is a subset of all theoretical populations I guess so it is still a sample. Nearly all the variables are dummy variables (a few are not) so non-linearity does not apply. There are tens of thousands of data points.

I am not sure if we should analyze for outliers or hetero or not.....