Inference on self-exciting jumps in prices and volatility using high-frequency measures

Worapree Maneesoonthorn, Catherine S. Forbes, Gael M. Martin

Research output: Contribution to journalArticleResearchpeer-review

7 Citations (Scopus)

Abstract

Dynamic jumps in the price and volatility of an asset are modelled using a joint Hawkes process in conjunction with a bivariate jump diffusion. A state-space representation is used to link observed returns, plus nonparametric measures of integrated volatility and price jumps, to the specified model components, with Bayesian inference conducted using a Markov chain Monte Carlo algorithm. An evaluation of marginal likelihoods for the proposed model relative to a large number of alternative models, including some that have featured in the literature, is provided. An extensive empirical investigation is undertaken using data on the S&P 500 market index over the 1996-2014 period, with substantial support for dynamic jump intensities-including in terms of predictive accuracy-documented.

Original languageEnglish
Pages (from-to)504-532
Number of pages29
JournalJournal of Applied Econometrics
Volume32
Issue number3
DOIs
Publication statusPublished - Apr 2017

Keywords

  • Dynamic price and volatility jumps
  • Stochastic volatility
  • Hawkes process
  • Nonolinear state space model
  • Bayesian Markov chain Monte Carlo
  • Global financial crisis

Cite this

@article{06879507a8384d4cab31ffaae897c474,
title = "Inference on self-exciting jumps in prices and volatility using high-frequency measures",
abstract = "Dynamic jumps in the price and volatility of an asset are modelled using a joint Hawkes process in conjunction with a bivariate jump diffusion. A state-space representation is used to link observed returns, plus nonparametric measures of integrated volatility and price jumps, to the specified model components, with Bayesian inference conducted using a Markov chain Monte Carlo algorithm. An evaluation of marginal likelihoods for the proposed model relative to a large number of alternative models, including some that have featured in the literature, is provided. An extensive empirical investigation is undertaken using data on the S&P 500 market index over the 1996-2014 period, with substantial support for dynamic jump intensities-including in terms of predictive accuracy-documented.",
keywords = "Dynamic price and volatility jumps, Stochastic volatility, Hawkes process, Nonolinear state space model, Bayesian Markov chain Monte Carlo, Global financial crisis",
author = "Worapree Maneesoonthorn and Forbes, {Catherine S.} and Martin, {Gael M.}",
year = "2017",
month = "4",
doi = "10.1002/jae.2547",
language = "English",
volume = "32",
pages = "504--532",
journal = "Journal of Applied Econometrics",
issn = "0883-7252",
publisher = "Wiley-Blackwell",
number = "3",

}

Inference on self-exciting jumps in prices and volatility using high-frequency measures. / Maneesoonthorn, Worapree; Forbes, Catherine S.; Martin, Gael M.

In: Journal of Applied Econometrics, Vol. 32, No. 3, 04.2017, p. 504-532.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Inference on self-exciting jumps in prices and volatility using high-frequency measures

AU - Maneesoonthorn, Worapree

AU - Forbes, Catherine S.

AU - Martin, Gael M.

PY - 2017/4

Y1 - 2017/4

N2 - Dynamic jumps in the price and volatility of an asset are modelled using a joint Hawkes process in conjunction with a bivariate jump diffusion. A state-space representation is used to link observed returns, plus nonparametric measures of integrated volatility and price jumps, to the specified model components, with Bayesian inference conducted using a Markov chain Monte Carlo algorithm. An evaluation of marginal likelihoods for the proposed model relative to a large number of alternative models, including some that have featured in the literature, is provided. An extensive empirical investigation is undertaken using data on the S&P 500 market index over the 1996-2014 period, with substantial support for dynamic jump intensities-including in terms of predictive accuracy-documented.

AB - Dynamic jumps in the price and volatility of an asset are modelled using a joint Hawkes process in conjunction with a bivariate jump diffusion. A state-space representation is used to link observed returns, plus nonparametric measures of integrated volatility and price jumps, to the specified model components, with Bayesian inference conducted using a Markov chain Monte Carlo algorithm. An evaluation of marginal likelihoods for the proposed model relative to a large number of alternative models, including some that have featured in the literature, is provided. An extensive empirical investigation is undertaken using data on the S&P 500 market index over the 1996-2014 period, with substantial support for dynamic jump intensities-including in terms of predictive accuracy-documented.

KW - Dynamic price and volatility jumps

KW - Stochastic volatility

KW - Hawkes process

KW - Nonolinear state space model

KW - Bayesian Markov chain Monte Carlo

KW - Global financial crisis

UR - http://www.scopus.com/inward/record.url?scp=84997018020&partnerID=8YFLogxK

U2 - 10.1002/jae.2547

DO - 10.1002/jae.2547

M3 - Article

AN - SCOPUS:84997018020

VL - 32

SP - 504

EP - 532

JO - Journal of Applied Econometrics

JF - Journal of Applied Econometrics

SN - 0883-7252

IS - 3

ER -