Skew selection for factor stochastic volatility models

Abstract

This paper proposes factor stochastic volatility models with skew error distributions. The generalized hyperbolic skew t-distribution is employed for commonfactor processes and idiosyncratic shocks. Using a Bayesian sparsity modeling strategy for the skewness parameter provides a parsimonious skew structure for possibly high-dimensional stochastic volatility models. Analyses of daily stock returns are provided. Empirical results show that the skewness is important for common-factor processes but less for idiosyncratic shocks. The sparse skew structure improves prediction and portfolio performance.

Stochastic Volatility Models with Skewness Selection

“…While Equation (1) implies that y t follows a symmetric distribution, Equation (11) indicates that y t comes from a potentially skewed distribution. Therefore, our model can be viewed as an extension of the stochastic volatility models with static skewness proposed in [7,8,17].…”

“…To the best of our knowledge, we are the first to introduce sparsity in the dynamic skewness framework. The closest paper to ours is [8], which uses the spike and slab prior of [30] to estimate the posterior probability of inclusion of static skewness.…”

“…This paper intersects and contributes to multiple areas. First, it contributes to the stochastic volatility literature by expanding the static skewness model of [7] to the dynamic case while also extending the sparsity-inducing scheme of [8] by allowing for dynamic skewness to shrink towards the static case. Second, it contributes to the toolbox of methods for recovering dynamic skewness.…”