Modeling Nonlinear Dependence Structure Using Logistic Smooth Transition Copula Model

Paravee Maneejuk, Woraphon Yamaka, Pisit Leeahtam

Abstract

This study introduces a new measure of dependence for financial studies in the context of nonlinear modelling, termed as the logistic smooth transition (LST) copula. The model is based on a bivariate copula incorporated with a threshold and smooth parameter. A Monte Carlo simulation shows that this dependence measure exhibits better performance than the classical copulas. Finally, this study applies the LST copula to measure the dependence structure between bond yields in advanced economies.