Examining the Leverage Effect, Dynamic Conditional Correlation, and Volatility Spillover Among Selected Indices of the Tehran Stock Exchange: Evidence from the ARMA-DCC-GJR-GARCH Model

In financial literature, there are two well-explored characteristics of volatility. The first pertains to the asymmetric reactions of volatility to positive and negative news, while the second involves the presence of volatility spillover (contagion) between markets and various financial assets. The asymmetric behavior of volatility refers to empirical evidence that a negative return shock causes a greater increase in volatility than a positive return shock of the same size. Also, concerning the asymmetric impact of news on stock volatility, two hypotheses, namely the leverage effect and volatility contagion, have been postulated. Accordingly, this study aims to explore the leverage effect, dynamic conditional correlation, and volatility spillover among ten selected indices of the Tehran Stock Exchange. These ten industry indices collectively constitute over 75 percent of the overall Tehran Stock Exchange index.

In this study, the ARMA (1,1) form was utilized to construct the mean model. Then, the GJR-GARCH model was used to check the leverage effects. Finally, the DCC-GARCH (1,1) framework was employed, which helped to deeply analyze the dynamic linkages in volatility among selected indices of the Tehran Stock Exchange. The daily return data of industry indices, comprising a total of 1117 observations, was utilized during the period from March 25, 2018, to November 16, 2022.

The result of the GJR coefficient, which was positive and significant for all return series -except for the Chemical and Oil Product Indexes- Indicates leverage effects exist. Also, the result of DCC (1,1) indicates the conditional correlation between all variables is positive and volatility spillover among them was strongly confirmed.

Financial markets, particularly the stock market, exhibit varied responses to positive and negative shocks, and these shocks impact the correlation between variables. For this reason, this research aimed to investigate the time frame during which the stock market underwent substantial fluctuations. This approach allowed for a more thorough and accurate examination of leverage effects, volatility spillover, and dynamic conditional correlation between returns. In the first half of 2019, despite the drop in the prices of commodities, oil, and the COVID-19 pandemic, the stock market experienced stunning growth, while investors were excited to buy regardless of the fundamental conditions of the companies. This enthusiasm to buy spread among stock market industries, but from the second half of 2019, the situation was completely reversed, and the market sold their shares at the slightest negative news. Hence, based on the results revealing the presence of leverage effect, dynamic conditional correlation, and volatility contagion, investors and portfolio managers can use these findings to mitigate risks and optimize their portfolios.