VALUE-AT-RISK (VAR) ANALYSIS AND LONG MEMORY: EVIDENCE FROM FIAPARCH IN ISTANBUL STOCK EXCHANGE (ISE)

In this study, i modeled istanbul stock exchange (ise) index returns using a number of symmetric and asymmetric conditional heteroscedasticity models including long memory models, namely garch, igarch, gjr-garch, aparch, figarch and fiaparch models. the accuracy of one-day-ahead value-at-risk (var) is examined based on the kupiec-lr test under the normal, student-t, and skewed student-t distributions. the results of arch class models show the existence of both leverage effect in stock exchange and fractional integration in conditional volatilities, which emphasizes the use of fiaparch model for ise. also the kupiec lr test based on in-sample and out-of-sample var confirms the superiority of fiaparch model. thus, student-t fiaparch modeling the leverage and long memory properties in ise index returns provides efficient var values. these findings would be helpful to the financial managers, investors, and regulators dealing with istanbul stock exchange.