文章基本信息

标题：Dynamic Quantile Linear Models: A Bayesian Approach
本地全文：下载
作者：Kelly C. M. Gonçalves ; Hélio S. Migon ; Leonardo S. Bastos 等
期刊名称：Bayesian Analysis
印刷版ISSN：1931-6690
电子版ISSN：1936-0975
出版年度：2020
卷号：15
期号：2
页码：335-362
DOI：10.1214/19-BA1156
语种：English
出版社：International Society for Bayesian Analysis
摘要：The paper introduces a new class of models, named dynamic quantile linear models, which combines dynamic linear models with distribution-free quantile regression producing a robust statistical method. Bayesian estimation for the dynamic quantile linear model is performed using an efficient Markov chain Monte Carlo algorithm. The paper also proposes a fast sequential procedure suited for high-dimensional predictive modeling with massive data, where the generating process is changing over time. The proposed model is evaluated using synthetic and well-known time series data. The model is also applied to predict annual incidence of tuberculosis in the state of Rio de Janeiro and compared with global targets set by the World Health Organization.
关键词：asymmetric Laplace distribution; Bayes linear; Bayesian quantile regression; dynamic models; Gibbs sampling