Open Access
June 2020 Dynamic Quantile Linear Models: A Bayesian Approach
Kelly C. M. Gonçalves, Hélio S. Migon, Leonardo S. Bastos
Bayesian Anal. 15(2): 335-362 (June 2020). DOI: 10.1214/19-BA1156


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.


Download Citation

Kelly C. M. Gonçalves. Hélio S. Migon. Leonardo S. Bastos. "Dynamic Quantile Linear Models: A Bayesian Approach." Bayesian Anal. 15 (2) 335 - 362, June 2020.


Published: June 2020
First available in Project Euclid: 26 April 2019

MathSciNet: MR4078717
Digital Object Identifier: 10.1214/19-BA1156

Keywords: asymmetric Laplace distribution , Bayes linear , Bayesian quantile regression , dynamic models , Gibbs sampling

Vol.15 • No. 2 • June 2020
Back to Top