ubes_a_1766470_sm0244.pdf (934.56 kB)
A Bayesian Quantile Time Series Model for Asset Returns
Version 2 2020-06-10, 21:48
Version 1 2020-05-12, 08:49
journal contribution
posted on 2020-06-10, 21:48 authored by Jim E. Griffin, Gelly MitrodimaWe consider jointly modeling a finite collection of quantiles over time. Formal Bayesian inference on quantiles is challenging since we need access to both the quantile function and the likelihood. We propose a flexible Bayesian time-varying transformation model, which allows the likelihood and the quantile function to be directly calculated. We derive conditions for stationarity, discuss suitable priors, and describe a Markov chain Monte Carlo algorithm for inference. We illustrate the usefulness of the model for estimation and forecasting on stock, index, and commodity returns.