استخدام الشبكات العصبية ذات الانحدار الذاتي غير الخطية NAR في التنبؤ بمعدل التضخم في الجزائر خلال الفترة 1990 إلى 2021

تاريخ الاستلام:23/11/2022 تاريخ القبول:20/03/2023

Abstract

Inflation is one of the most serious economic problems facing most economies, whether developed or developing countries, which require central banks to implement policies to prevent high price fluctuations and long periods of inflation. This study uses both nonlinear autoregressive neural network (NARNN) and autoregressive integrated moving average (ARIMA) models to forecast Algeria's inflation rate. The empirical analysis is carried out using annual time series data from 1990 to 2021. The mean squared errors (MSE) and R squared are used to compare the outcomes of the two models. According to the results obtained the NARNN model outperformed the ARIMA model.