Using apgarch/avgarch models Gaussian and non-Gaussian for modeling volatility exchange rate

Document Type : Research Paper


Department of Statistics, University of Baghdad, Baghdad, Iraq


This paper aims to measure the effect of the volatility on the daily closing price for the (Iraqi dinar against US dollar) from (21 July 2011) until (21  July 2021) using the models of asymmetric general autoregressive conditional heterogeneity (APGARCH and AVGARCH). The parameter estimated by Maximum  Likelihood Estimation method and the error term assumed two distributional (General error distribution and Student's t distribution), the results showed that the APGARCH(1,2) with error term distributed (Student's t) distribution is the best model for the return series of the (IQ/USD) exchange rate to get the lowest value according to the information criteria for determining ranks (AIC, BIC) in addition to the presence of the asymmetric effect of the leverage, and this is evidence that negative shocks affect volatility more than positive shocks (the impact of the positive shocks is less than the impact of the negative shocks).