This project has been realised in the context of the Regression Analysis course of the KULeuven.
It is divided into three parts and datasets:
-The hospital data, on which the best fitting regression is found using quadratic terms, partial residual plots, variable transformation, weighted least squares and Ridge regression
-The fossil data, on which non parametric regressions are fitted
-The fuel data, on which Robust Least Trimmed Squares is fitted
Every analysis looks at residual plots and checks assumptions.