This repository conatins a complication of labs, tutorials and personal mini-projects from UNSW "linear models", which focusses on using R-Studio to conduct analysis on data using linear rgeression. Examining simple linear regression, focusing on parameter estimation using least squares, the properties of estimators, error variance estimation, and hypothesis testing.
Extends this to the general linear model, covering least squares estimation, maximum likelihood estimation, hypothesis testing, and multicollinearity. Model selection methods are also explored, including cross-validation, PRESS residuals, and variable selection procedures. Additionally, the text addresses residual analysis and diagnostics, including residual plots, outlier detection, and influence measures. It also covers categorical predictors, less than full rank linear models, and logistic regression, emphasizing hypothesis testing in these contexts.