Covers Generalized Linear Models, likelihood, deviance, link functions, sampling methods, Poisson regression, over-dispersion, and alternative regression models.
Covers regression diagnostics for linear models, emphasizing the importance of checking assumptions and identifying outliers and influential observations.
Explores Generalized Linear Models for non-Gaussian data, covering interpretation of natural link function, MLE asymptotic normality, deviance measures, residuals, and logistic regression.
Explores special examples of Generalized Linear Models, covering logistic regression, count data models, separation issues, and nonparametric relationships.
Explores the application of Maximum Likelihood Estimation in binary choice models, covering probit and logit models, latent variable representation, and specification tests.
Explores the provable benefits of overparameterization in model compression, emphasizing the efficiency of deep neural networks and the importance of retraining for improved performance.
