Gaussian Process Regression Analysis for Functional Data

ISBN-13: 9781439837733 / Angielski / Twarda / 2011 / 216 str.

Jian Qing Shi; Taeryon Choi

Opis

A Gaussian process is a type of stochastic process that is particularly useful for Bayesian inference of complex problems. This book presents an overview of Gaussian process regression, all of the necessary theory, and real-world applications from the fields of medicine and engineering. It covers the required background in regression and examines key aspects of the modeling process, including model selection as well as more advanced topics, such as mixture models and kernel-based methods. For implementing the methods discussed in the text, the authors provide ad hoc software for download on their website.

Spis treści

Introduction. Bayesian Nonlinear Regression with Gaussian Process Priors. Inference and Computation for Gaussian Process Regression Model. Covariance Function and Model Selection. Functional Regression Analysis. Mixture Models and Curve Clustering. Generalized Gaussian Process Regression for Non-Gaussian Functional Data. Some Other Related Models. Appendices. Bibliography. Index.