Analyzing Exchange Rate Volatility: A Comparative Study of ARCH and GARCH Methods

Authors : Mesut Fenkli, Ayşe Nur Çırak, Doğan Uysal

Pages : 121-142

Doi:10.52791/aksarayiibd.1370072

View : 160 | Download : 293

Publication Date : 2024-09-27

Article Type : Research Paper

Abstract :The volatility experienced in exchange rates concerns companies and investors, especially governments, and requires these users to be informed about the fluctuations experienced. In this study, the volatility in weekly foreign exchange sales prices of the Central Bank of the Republic of Turkey over the US Dollar and European Union Euro currencies, which consists of 1232 observations, between 1999 (based on the transition year of the European Union to the common currency of the European Union) and 2022 for Turkey has been examined. In the application of the study, autoregressive conditional variance (ARCH) and generalized autoregressive conditional variance (GARCH) methods, which are frequently used in time series, were used. Models of these methods are estimated separately for both exchange rates. As a result of the model predictions, it was determined that the GARCH (1,1) model was successful in explaining the volatility in both exchange rates. As a result, it has been decided that the volatility experienced in Dollar and Euro exchange rates between 1999-2022 in Turkey (over the past prices of the exchange rates) can be estimated using the GARCH model and has the GARCH effect.
Keywords : döviz kuru, zaman serisi, ARCH ve GARCH modeller