Bayesian segmented regression and application

Regression models, which typically model the relationship between variables or provide prediction, are very important tools in data science and machine learning! Regression with Multiple Change Points (or switch points, break points, broken) is often called segmented regression and has a wide application, in particular, when the independent variables cluster into different groups or exhibit different relationships between the variables in these regions. The boundaries between the segments are change points, which are often unknown in practice. Bayesian inference is a method of statistical inference in which Bayes' theorem is used to update the probability for a hypothesis as more evidence or information becomes available.

This project aims to develop different algorithms, efficient algorithms under Bayesian inference for those regression problems, including unknown number of change-point estimation, prior selection, likelihood structure and posterior inference, high-dimensional data analysis and modern variable selection, model misspecification and robust inference and test, segment trees under Machine Learning.

Meet the Principal Investigator(s) for the project

Professor Keming Yu - Keming Yu – Chair in Statistics Impact champion of REF – in UOA (Mathematical Sciences) Keming joined Brunel University London in 2005. Before that he held posts at various institutions, including University of Plymouth, Lancaster University and the Open University. Keming got his first degree in Mathematics and MSc in Statistics from universities in China and got his PhD in Statistics from The Open University, Milton Keynes.

Project last modified 28/01/2021