Effects of additive outliers on asymmetric garch models

This study examines effects of additive outliers on asymmetric generalized autoregressive conditional heteroscedastic (garch) models. the outlier’s detection and correction method is developed for asymmetric garch models. we focus on the estimates of parameters and forecasts of out-of-sample volatility when data is contaminated with outliers. the performance of the proposed method is assessed through monte carlo simulations and an application to real data. our results show that the asymmetric garch model fitted to outlier corrected data yields better estimates of both parameters and volatility forecasts than the asymmetric models for original returns. © 2017 pakistan journal of statistics.