Project Part of the Course METHODS AND MODELS OF STATISTICAL INFERENCE (A.Y. 2023/2024)
Authors: Maria Chiara Menicucci, Leonardo Pascotto, Alessandro Pedone, Arianna Perotti
Content:
- LaTeX presentation
- dataset with code for creating the .txt file
- code for data analysis
Dataset: OECD PISA 2022 survey data available on the OECD website
Tools Used:
- data visualization
- non-parametric ANOVA (Kruskal-Wallis and Dunn's test)
- removal of influential points (outliers and leverages)
- variable selection (stepwise backward and forward selection with BIC and AIC)
- model validation (homoscedasticity, normality test)
- VIF (Variance Inflation Factor) and GVIF (generalized VIF)
- cross-validation with K-fold and leave-one-out
- prediction and confidence intervals