No Cover Image

Journal article 671 views

A new Bayesian approach to quantile autoregressive time series model estimation and forecasting / Yuzhi Cai, Julian Stander, Neville Davies

Journal of Time Series Analysis, Volume: 33, Issue: 4, Pages: 684 - 698

Swansea University Author: Yuzhi Cai

Full text not available from this repository: check for access using links below.

Abstract

This paper proposes a Bayesian approach to quantile autoregressive (QAR) time series model estimation andforecasting. We establish that the joint posterior distribution of the model parameters and future values is welldefined. The associated Markov chain Monte Carlo algorithm for parameter estimatio...

Full description

Published in: Journal of Time Series Analysis
ISSN: 0143-9782
Published: wileyonlinelibrary.com 2012
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa11711
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: This paper proposes a Bayesian approach to quantile autoregressive (QAR) time series model estimation andforecasting. We establish that the joint posterior distribution of the model parameters and future values is welldefined. The associated Markov chain Monte Carlo algorithm for parameter estimation and forecasting convergesto the posterior distribution quickly. We also present a combining forecasts technique to produce more accurateout-of-sample forecasts by using a weighted sequence of fitted QAR models. A moving window method to checkthe quality of the estimated conditional quantiles is developed. We verify our methodology using simulationstudies and then apply it to currency exchange rate data. The results obtained show that an unequally weightedcombining method performs better than other forecasting methodology.
Keywords: Combining forecasts; MCMC; quantile modelling; quantile forecasting; predictive density functions
College: School of Management
Issue: 4
Start Page: 684
End Page: 698