Sequential Estimation of Multivariate Factor Stochastic Volatility Models

TL;DR

This paper introduces a simple, two-step estimation method for multivariate factor stochastic volatility models that reduces computational complexity and performs well in simulations and real data applications.

Contribution

The paper proposes a novel, efficient estimation approach for multivariate factor stochastic volatility models, combining maximum likelihood and method of moments, improving computational feasibility.

Findings

Accurately estimates model parameters in simulations.

Demonstrates computational advantages over existing methods.

Effective on real data with up to 148 dimensions.

Abstract

We provide a simple method to estimate the parameters of multivariate stochastic volatility models with latent factor structures. These models are very useful as they alleviate the standard curse of dimensionality, allowing the number of parameters to increase only linearly with the number of the return series. Although theoretically very appealing, these models have only found limited practical application due to huge computational burdens. Our estimation method is simple in implementation as it consists of two steps: first, we estimate the loadings and the unconditional variances by maximum likelihood, and then we use the efficient method of moments to estimate the parameters of the stochastic volatility structure with GARCH as an auxiliary model. In a comprehensive Monte Carlo study we show the good performance of our method to estimate the parameters of interest accurately. The simulation study and an application to real vectors of daily returns of dimensions up to 148 show the method's computation advantage over the existing estimation procedures.