Generalized changepoint regression

Changepoint detection is widely used in modern data streams, but current regression based changepoint methods are limited in flexibility. Existing approaches typically assume that when a changepoint occurs, it affects all regression variables at the same time. This makes it difficult to model realistic scenarios where some components remain fixed while others change. This project aims to develop an open source R package implementing recent work on generalised changepoint regression, where users can specify both fixed and changing regression covariates. The approach builds on existing research level code, which will be refactored into a generic framework function capable of handling a range of regression models, including time series settings. The project will deliver a fully functioning R package with user-facing functions and supporting tools such as summary, print and plotting methods, designed to be consistent with existing changepoint packages. It will also include extensive testing, simulation functions for generating data, and clear documentation, with the goal of producing a CRAN ready package. Should time allow adding mixed effects or non-parametric regression models would be included. Overall, this project will provide a flexible and practical tool that addresses a gap in current changepoint methodology and will be useful for applied users.