Modeling exchange rate volatility, using Univariate Generalized Autoregressive conditionally Hetroscedastic type models: evidence from Afghanistan

In  this study, an attempt  is made  to examine  the performance of GARCH  family models  (including  symmetric GARCH,GARCH-M, and asymmetric EGARCH  models)  in Forecasting the volatility behavior of Afghanistan foreign exchange rate. Daily foreign exchange rates of Afghani with USD data, ranging from 2018/09/01 to 2019/10/16 are used. Theoretically,  the  first  order  autoregressive  behavior  of  the  foreign exchange  rate  was evidenced  in GARCH, GARCH-M and E-GARCH models while  GARCH, GARCH-M and E-GARCH models support that previous day foreign exchange rate  affected  the  current  day  exchange  rate.  Based on the comparison of the above models, found the GARCH (1,1) is the best model to explain the volatility of the return on the exchange of AFN with US Dollar