Statistical investigation of how sandwich components affect ant attraction, based on a full factorial design with ANOVA and Poisson regression.

Fast Best-Subset Selection Library

Poisson pseudo-likelihood regression with multiple levels of fixed effects

Regression Models for Epidemiology

Coursework, Stata code, and notes for PBHS 32700: Biostatistical Methods (Spring 2024, University of Chicago). Topics include contingency tables, logistic regression, Poisson and negative binomial models, and survival analysis using Kaplan-Meier, Cox, and parametric models. The course emphasizes categorical and time-to-event analysis using Stata.

Coursework and Stata code for PBHS 32410: Regression Analysis for Health and Social Research (Winter 2024, University of Chicago). Topics include linear regression, multiple regression, interaction effects, model diagnostics, Poisson and logistic regression, and the application of generalized linear models in public health data.

Coursework, Stata code, and notes for PBHS 31001: Epidemiologic Methods (Winter 2024, University of Chicago). Topics include bias, confounding, effect modification, cohort and case-control study design, and logistic/Poisson regression. The course emphasizes observational study design and practical applications using Stata.

Fast Change Point Detection in R

BAS R package for Bayesian Model Averaging and Variable Selection

Latent gaussian processes for zero inflated count data.

Extended Elo rating system implementation based on the equivalence with logistic regression.

a generalized linear models python library implementation from scratch

This repository contains implementations of advanced regression methods, including ordinary least squares, Poisson regression, and locally weighted regression. It also explores bias-variance decomposition for regularized mean estimators. The analysis is conducted on the Capital Bikesharing dataset using Python.

Data Science Project (Regression for Numeric Data)

A bot that provides soccer predictions using Poisson regression

The project aims at using quasi-experimental designs to estimate the impact of policy reducing speed limits in major cities, on the number of road accidents. It is based on Bayesian hierarchical modelling and Poisson regression

Benchopt benchmark for GLMs

This project aims to develop a Poisson regression model to predict the number of rings in a dataset. Poisson regression is a type of generalized linear model used for modeling count data, where the target variable represents the number of occurrences of an event within a given time or space. In this case, the target variable is the count of rings.

R package 'bpr' to perform posterior inference for Bayesian Poisson regression

A series of Statistical Modelling assignments with the use of R. Applications of Linear, Polynomial, Logistic and Poisson Regression in various datasets