Modeling the Dependency in the Turkish Stock Market via the Dynamic Vine-Garch Model

Evkaya O., Poyraz G., Yıldırım Külekci B., Gür İ.

25th (2022) International Congress on Insurance: Mathematics and Economics, Guangzhou, Çin, 12 - 15 Temmuz 2022, ss.62

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: Guangzhou
  • Basıldığı Ülke: Çin
  • Sayfa Sayıları: ss.62
  • Orta Doğu Teknik Üniversitesi Adresli: Hayır

Özet

The dependence on financial instruments has always been an interesting topic considering their economic and political effects. To measure the change in such instruments over time and interaction between them, the stock prices are widely considered indicators in the literature. While the dynamic portfolio with a moving window approach is proposed for capturing the changing structure, the Vine-copula is used for the highdimensional dependence. In this study, we analyze ISE100 stocks in the pre-, during-, and post-global financial crisis (GFC) periods by first dividing the stocks into subsectors. For each subsector and time period, ARMA-GARCH models are implemented and compared for different innovations. Thereafter, the dependence structure for each subsector over different periods is modeled with the R-vine copula model. Additionally, Value at Risk (VaR) and expected shortfall (ES) risk measures are computed by Monte Carlo simulation by assuming an equally weighted portfolio, constructed by the selected sector leaders. The applied model offers more consistent and sensitive results with the inclusion of Vine-copula as well as suitable innovations distribution. The primary findings of the study show that the proposed Vine-GARCH Model gives more accurate results in capturing portfolio risk compared to the classical approaches.