التنبؤ باسعار صرف العملات الأجنبية في سوق الفوركس

Abstract

This study aims to identify the foreign exchange market, and apply time -series models in forecasting the euro-dollar exchange rate, based on the daily data of closing prices for the period from 03-01-2000 to 03-12-2020 within the forex market. Statistical tests were applied to study stationarity and detect the presence of heteroskedasticity, then estimate a set of models and compare among them using information criteria, where it was found that the best model is ARIMA(0.1.1) - GARCH(2.4) In order to find out the validity of the model for prediction, the prediction within the sample was used , the comparison between the ARIMA(0.1.1) model and the ARIMA(0.1.1) -GARCH(2.4) model using the prediction accuracy criteria showed that the linear model ARIMA(0.1.1) is the best for prediction based on the values of MSE- RMSE-MAE-MAPE