The Sparse Multivariate Method of Simulated Quantiles

TL;DR

This paper extends the method of simulated quantiles to a multivariate framework with a sparse estimator for the scale matrix, enabling effective analysis of complex distributions and high-dimensional data.

Contribution

It introduces a multivariate version of MSQ with a sparse estimator using SCAD penalty, establishing its theoretical properties and demonstrating its effectiveness on synthetic and real data.

Findings

The sparse-MMSQ estimator is consistent and asymptotically normal.

It achieves oracle properties under mild conditions.

The method outperforms existing approaches on stable distribution data.

Abstract

In this paper the method of simulated quantiles (MSQ) of Dominicy and Veredas (2013) and Dominick et al. (2013) is extended to a general multivariate framework (MMSQ) and to provide a sparse estimator of the scale matrix (sparse-MMSQ). The MSQ, like alternative likelihood-free procedures, is based on the minimisation of the distance between appropriate statistics evaluated on the true and synthetic data simulated from the postulated model. Those statistics are functions of the quantiles providing an effective way to deal with distributions that do not admit moments of any order like the $\alpha$-Stable or the Tukey lambda distribution. The lack of a natural ordering represents the major challenge for the extension of the method to the multivariate framework. Here, we rely on the notion of projectional quantile recently introduced by Hallin etal. (2010) and Kong Mizera (2012). We establish consistency and asymptotic normality of the proposed estimator. The smoothly clipped absolute deviation (SCAD) $\ell_1$--penalty of Fan and Li (2001) is then introduced into the MMSQ objective function in order to achieve sparse estimation of the scaling matrix which is the major responsible for the curse of dimensionality problem. We extend the asymptotic theory and we show that the sparse-MMSQ estimator enjoys the oracle properties under mild regularity conditions. The method is illustrated and its effectiveness is tested using several synthetic datasets simulated from the Elliptical Stable distribution (ESD) for which alternative methods are recognised to perform poorly. The method is then applied to build a new network-based systemic risk measurement framework. The proposed methodology to build the network relies on a new systemic risk measure and on a parametric test of statistical dominance.