References & Citations
Statistics > Methodology
Title: Dynamic quantile linear model: a Bayesian approach
(Submitted on 1 Nov 2017 (this version), latest version 18 Feb 2018 (v2))
Abstract: A new class of models, named dynamic quantile linear models, is presented. It combines the dynamic linear models with the distribution free quantile regression producing a robust statistical method. Bayesian inference for the dynamic quantile linear models can be done using an efficient Markov chain Monte Carlo algorithm. A fast sequential procedure suited for high-dimensional predictive modeling applications with massive data, in which the generating process is itself changing overtime, is also proposed. The proposed model is evaluated using synthetic and well-known time series data. The model is also applied to predict annually incidence rate of Tuberculosis in Rio de Janeiro for the next years, opposing it to the the Global Tuberculosis strategy targets.
Submission history
From: Kelly Cristina Mota Goncalves [view email][v1] Wed, 1 Nov 2017 02:01:04 GMT (392kb,D)
[v2] Sun, 18 Feb 2018 18:43:18 GMT (389kb,D)
Link back to: arXiv, form interface, contact.