Time-Varying Mixture Copula Models with Copula Selection

Abstract

Modeling the joint tails of multiple financial time series has many important implications for risk management. Classical models for dependence often encounter a lack of fit in the joint tails, calling for additional flexibility. This paper introduces a new semiparametric time-varying mixture copula model, in which both weights and dependence parameters are deterministic and unspecified functions of time. We propose penalized time-varying mixture copula models with group smoothly clipped absolute deviation penalty functions to do the estimation and copula selection simultaneously. Monte Carlo simulation results suggest that the shrinkage estimation procedure performs well in selecting and estimating both constant and time-varying mixture copula models. Using the proposed model and method, we analyze the evolution of the dependence among four international stock markets, and find substantial changes in the levels and patterns of the dependence, in particular around crisis periods.

Economic Policy Uncertainty: Cross-Country Linkages and Spillover Effects on Economic Development in Some Belt and Road Countries

“…Finally, we would like to note that to avoid a possible misspecification of a chosen copula, one can follow the copula selection approach proposed in Cai and Wang [36] . Also, note that the aforementioned specification can be possibly updated as a dynamic form using the newly proposed techniques as in Yang, et al [38] and Liu, et al [39] .…”

A review of advances in imaging methodology in fluorescence molecular tomography

“…More exactly, the fluorescent source can be selected as a group variable to achieve morphological reconstruction. So far, several group sparsity penalty functions have been proposed and utilized (Jiang et al 2019a, 2019b, Jiang et al 2020, mainly including the L 2,1 norm model (Jiang et al 2019b), the fusion LASSO model (Jiang et al 2019a, Jiang et al 2020 and the group smoothly clipped absolute deviation (Yang et al 2018).…”

“…In 2019, our group (Guo et al 2019) proposed a 3D-En-Decoder deep learning framework for 3D FMT reconstruction (Yang et al 2018). As shown in figure 5(a), it is a typical encoder-decoder network structure framework that establishes a nonlinear mapping relationship between internal fluorescence sources and boundary fluorescence measurements.…”