文章基本信息

标题：Variational Bayes for Functional Data Registration, Smoothing, and Prediction
本地全文：下载
作者：Cecilia Earls ; Giles Hooker
期刊名称：Bayesian Analysis
印刷版ISSN：1931-6690
电子版ISSN：1936-0975
出版年度：2017
卷号：12
期号：2
页码：557-582
DOI：10.1214/16-BA1013
语种：English
出版社：International Society for Bayesian Analysis
摘要：We propose a model for functional data registration that extends current inferential capabilities for unregistered data by providing a flexible probabilistic framework that 1) allows for functional prediction in the context of registration and 2) can be adapted to include smoothing and registration in one model. The proposed inferential framework is a Bayesian hierarchical model where the registered functions are modeled as Gaussian processes. To address the computational demands of inference in high-dimensional Bayesian models, we propose an adapted form of the variational Bayes algorithm for approximate inference that performs similarly to Markov Chain Monte Carlo (MCMC) sampling methods for well-defined problems. The efficiency of the adapted variational Bayes (AVB) algorithm allows variability in a predicted registered, warping, and unregistered function to be depicted separately via bootstrapping. Temperature data related to the El-Niño phenomenon is used to demonstrate the unique inferential capabilities for prediction provided by this model.