COMPARISON OF ARIMA MODEL AND ANN MEASURES IN FORECASTING REGIONAL MACROECONOMIC INDICATORS

Abstrak

In this article, the process of modeling and forecasting the volume of the gross regional product of the Surkhandarya region was carried out. The forecasting efficiency of the ARIMA and ANN models is also compared. As a result, the best ARIMA model in forecasting is ARIMA (1, 0, 1)(0, 1, 0)4, and according to ANN, the model of the output neuron MLP (4-7-1) turned out to be the best. In addition, the volume of the gross regional product is predicted until 2025.

Kalit soʻzlar GRP volume, ARIMA model, ANN model, Box-Jenkins method, Akaike information criterion, Schwartz information criterion, forecasting