Developing Non-linear Dynamic Model to Estimate Value at Risk, Considering the Effects of Asymmetric News: Evidence from Tehran Stock Exchange

摘要

Empirical studies concluded the existence of stronger correlation among the major losses compared to the major profits in financial markets. This phenomenon makes symmetric distributions inefficient for modeling multivariate distributions and estimating portfolio's risk perfectly. Copula theory is an appropriate tool in order to model multivariate distributions which use marginal distribution and hires defined asset's correlation to describe complex correlation structure such as non-linear one. Therefore, this study calculated the risk of a portfolio including five-industry indexes in Tehran Stock Exchange Market with application of Value at Risk measure. In this regard, marginal distribution of each return series was estimated using GARCH and GJR models, and also correlation structure of assets was determined by implementation of DCC model. Subsequently, joint distribution of asset's portfolio is achieved and finally VaR of equal weighted portfolio for each asset is calculated. The result of kupiec test illustrated that the proposed model calculated VaR efficiently, and also the amount of VaR calculated by t-Copula is less than Gaussian-Copula in 99 and 95 percent of the significant level.