Multivariate Regime Switching Model with Flexible Threshold Variable

37 Pages Posted: 16 Jan 2014 Last revised: 8 Feb 2015

Date Written: November 17, 2014

Abstract

This paper proposes a novel multivariate regime switching model that allows the threshold variable to be a linear combination of covariates with unknown coefficients: the model is likely to be more suitable to analyze time series of data in which regimes dynamics are driven by multiple covariates rather than by just one single threshold variable. The paper considers least squares estimation of the model and it proposes a test for the number of regimes based on theoretical results from multivariate statistics. Finite sample results from Monte Carlo analysis strengthen the methodological contribution of the paper. An application to measuring regime-specific cross-sectional dependence in the U.S. stock market illustrates the usefulness of the proposed model for applied work.

Keywords: Multivariate Threshold Model, Flexible Threshold Variable, Regime-Specific Cross-Sectional Dependence, Stock Returns

JEL Classification: C31, C32, G11

Suggested Citation

Massacci, Daniele, Multivariate Regime Switching Model with Flexible Threshold Variable (November 17, 2014). Available at SSRN: https://ssrn.com/abstract=2377220 or http://dx.doi.org/10.2139/ssrn.2377220

Daniele Massacci (Contact Author)

King's College London ( email )

United Kingdom

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
118
Abstract Views
619
rank
260,067
PlumX Metrics