Estimating Stable Factor Models by Indirect Inference

Financial returns exhibit common behavior described at best by factor models, but also fat tails, which may be captured by stable distributions. This paper concentrates on estimating factor models with multivariate stable distributed and independent latent factors and idiosyncratic noises under the assumption of time constant distribution (static factor models) or time-varying conditional distribution (GARCH factor models). While the simulation from such a distribution is straightforward, the estimation encounters difficulties. These difficulties are overcome in this paper by implementing the indirect inference estimation method with the multivariate Student’s t as the auxiliary distribution.