Download this Paper Open PDF in Browser

Estimating Stable Latent Factor Models by Indirect Inference

37 Pages Posted: 8 Jan 2015 Last revised: 13 Apr 2016

Giorgio Calzolari

Universita di Firenze - Dipartimento di Statistica

Roxana Halbleib-Chiriac

University of Konstanz

Date Written: February 6, 2016

Abstract

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.

Keywords: Symmetric Multivariate alpha-stable Distribution, Latent Factor Models, Indirect Inference, Multivariate Student’s t Distribution, Discrete Spectral Measures, GARCH Models

Suggested Citation

Calzolari, Giorgio and Halbleib-Chiriac, Roxana, Estimating Stable Latent Factor Models by Indirect Inference (February 6, 2016). Available at SSRN: https://ssrn.com/abstract=2546302 or http://dx.doi.org/10.2139/ssrn.2546302

Giorgio Calzolari

Universita di Firenze - Dipartimento di Statistica ( email )

Viale Morgagni, 59
50134 Firenze
Italy
+39 055 4237 217 (Phone)
+39 055 4223 560 (Fax)

Roxana Halbleib-Chiriac (Contact Author)

University of Konstanz ( email )

Universitaetsstr. 10
Box: D 124
78457 Konstanz
Germany

HOME PAGE: http://econometrics.wiwi.uni-konstanz.de/staff/halbleib.htm

Paper statistics

Downloads
72
Rank
275,511
Abstract Views
423