The project explores a number of logistic regression models for binary classification. Specifically, the frequentist approach to logistic regression is compared against a number of Bayesian models including:
- Logistic regression using the Laplace approximation of the posterior distribution
- Bayesian Hierarchical models
Inference for the Bayesian models is performed using PyStan.