Bayesian stochastic volatility models for high-frequency data
Applications
2016-02-02 v1 Methodology
Abstract
We formulate a discrete-time Bayesian stochastic volatility model for high-frequency stock-market data that directly accounts for microstructure noise, and outline a Markov chain Monte Carlo algorithm for parameter estimation. The methods described in this paper are designed to be coherent across all sampling timescales, with the goal of estimating the latent log-volatility signal from data collected at arbitrarily short sampling periods. In keeping with this goal, we carefully develop a method for eliciting priors. The empirical results derived from both simulated and real data show that directly accounting for microstructure in a state-space formulation allows for well-calibrated estimates of the log-volatility process driving prices.
Cite
@article{arxiv.1602.00202,
title = {Bayesian stochastic volatility models for high-frequency data},
author = {Georgi Dinolov and Abel Rodriguez and Hongyun Wang},
journal= {arXiv preprint arXiv:1602.00202},
year = {2016}
}