Multivariate Garch with dynamic beta

TL;DR

This paper introduces a simplified multivariate GARCH model with a factor structure that uses only six parameters, enabling dynamic beta estimation and efficient covariance matrix inversion for large markets.

Contribution

It proposes a novel factor-based multivariate GARCH model with dynamic betas and analytical inverse covariance, improving scalability and interpretability.

Findings

Model is competitive with existing GARCH models on S&P 500 data.

Successfully captures market transition in 2006.

Provides insights into the leverage effect in financial markets.

Abstract

We investigate a solution for the problems related to the application of multivariate GARCH models to markets with a large number of stocks by restricting the form of the conditional covariance matrix. The model is a factor model and uses only six free GARCH parameters. One factor can be interpreted as the market component, the remaining factors are equal. This allow the analytical calculation of the inverse covariance matrix. The time-dependence of the factors enables the determination of dynamical beta coefficients. We compare the results from our model with the results of other GARCH models for the daily returns from the S\&P500 market and find that they are competitive. As applications we use the daily values of beta coefficients to confirm a transition of the market in 2006. Furthermore we discuss the relationship of our model with the leverage effect.