Analysis of Volatility Structure Using Conditional Heteroscedasticity Models: The Case of Participation 30 and Participation 50 Indexes

Abstract

This study aims to determine the most suitable model that explains the volatility structures of the Participation 30 and Participation 50 indexes traded in Borsa Istanbul (BIST). In order to determine the volatility structures of the aforementioned indexes; ARCH, GARCH, EGARCH, TGARCH, and PARCH models, which are called conditional heteroscedasticity models, are included in the analysis. As a result of the study, in which the weekly closing prices of the 2015-2020 period were used, it was determined that the most suitable volatility model for both indexes was the EGARCH (1,1) model and that the negative shocks in the indexes were more effective than the positive shocks. Another finding was that the effect of a shock in the indexes lasted approximately 32 days in the Participation 30 index and 28 days in the Participation 50 index.