Summary points
In lecture 10, we discussed
Common issues that could happen while checking model assumptions, with some suggestions on how to rectify these issues. Please note that this is not an exhaustive list and please treat each model fit individually.
Understanding possible outliers via
\(\mbox{Cook's Distance} \propto \sum_{j=1}^{n}(\hat{y}_j - \hat{y}_{j(i)})^2.\)
Leverage, corresponding to the diagonal entries of the hat matrix \(\boldsymbol{X}(\boldsymbol{X^T}\boldsymbol{X})^{-1}\boldsymbol{X}^T\)