Performance Comparison of tcopula GARCH-LVaR with GARCH-VaR To optimize the portfolio in the Tehran Stock Exchange

Subject Areas : Stock Exchange

Gholam Reza Taghizadegan ¹ (Department of Financial Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran)
, Gholamreza Zomorodian ² (Department of Financial Management, central Tehran Branch, Islamic Azad University. Tehran, Iran and member of Modern Financial Risk Research Group)
rasoul saadi, ³ (Department of Financial Management, central Tehran Branch, Islamic Azad University. Tehran, Iran)
mirfeyz Fallah ⁴ (Department of Financial Management, central Tehran Branch, Islamic Azad University. Tehran, Iran and member of Modern Financial Risk Research Group)

Received: 2022-11-15 Accepted : 2023-01-17 Published : 2023-09-23

Keywords: Portfolio optimization, Tehran Stock Exchange, tcopula, LVaR, Dynamic Conditional Correlation(DCC),

Abstract :

The aim of this research is to compare the performance of the value-at-risk model with the liquidity-t-copula approach with dynamic conditional correlation (t-copula-GARCH-LVaR) with the value-at-risk (VaR) model to optimize the portfolio in the Tehran Stock Exchange. In the current research, in order to test the desired hypotheses, the period is between 2001 and 2021. All the variables used in this research on a quantitative scale and observations in the form of time series are the daily logarithmic returns of 40 stock market indices, including 39 industry indices and one index of fixed-income bonds from the beginning of September 2011 to the end of July 2021. In this research, to perform the final analysis, all the calculations required for this research were done using the open-source software R 4.2.1. Our results showed that the t-copula-GARCH-LVaR optimization model performs better according to the Sharpe criterion based on Mann–Whitney U test at the 95% test level.