What are the best way to know which diagnostics check is best for the dataset you're using or the regression model you're using?
My lecturer has spoken about all these listed below, o is it okay to use any of these because they all provide different insight?
-
5 Checking for normally distributed residuals
-
6 Checking constant residual variance
-
7 Checking for relationships between residuals and predicted outcome or predictors
-
8 Checking linearity
- 8.1 What should be linear in a linear model?
- 8.2 Checking for linearity
-
9 Checking influence: leave-one-out analyses
- 10 Checking the variance inflation factors (VIFs)