Maximum Likelihood Estimation of the Markov-Switching GARCH Model

Computational Statistics & Data Analysis (DOI: 10.1016/j.csda.2013.01.026)

32 Pages Posted: 29 Dec 2016

See all articles by Maciej Augustyniak

Maciej Augustyniak

University of Montreal - Department of Mathematics and Statistics

Date Written: April 20, 2013

Abstract

The Markov-switching GARCH model offers rich dynamics to model financial data. Estimating this path dependent model is a challenging task because exact computation of the likelihood is infeasible in practice. This difficulty led to estimation procedures either based on a simplification of the model or not dependent on the likelihood. There is no method available to obtain the maximum likelihood estimator without resorting to a modification of the model. A novel approach is developed based on both the Monte Carlo expectation-maximization algorithm and importance sampling to calculate the maximum likelihood estimator and asymptotic variance-covariance matrix of the Markov-switching GARCH model. Practical implementation of the proposed algorithm is discussed and its effectiveness is demonstrated in simulation and empirical studies.

Keywords: Markov-switching, GARCH, EM algorithm, importance sampling

JEL Classification: C13, C51, C63

Suggested Citation

Augustyniak, Maciej, Maximum Likelihood Estimation of the Markov-Switching GARCH Model (April 20, 2013). Computational Statistics & Data Analysis (DOI: 10.1016/j.csda.2013.01.026), Available at SSRN: https://ssrn.com/abstract=2891147

Maciej Augustyniak (Contact Author)

University of Montreal - Department of Mathematics and Statistics ( email )

C.P. 6128, succursale Centre-ville
Montreal, Quebec H3C 3J7
Canada

HOME PAGE: http://dms.umontreal.ca/

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
185
Abstract Views
817
Rank
294,831
PlumX Metrics