Regression
models make assumptions that need to be reasonably satisfied
to make the model valid, i.e. that it gives reliable predictions,
and correct inferences about the importance of the potential
predictors. The three main types of assumptions (Harrell
et al., 1985) to explore are: