In this research work, autoregressive distribution lag (ARDL) models have been developed that assess the long-term and short-term influence of factors influencing the sustainable socio-economic development of the region on the volume of GRP per capita. Based on the developed models, the priority levels of factors were determined.
This study intends to forecast the agricultural commodities in Uzbekistan. The data were obtained from National Statistical Committee of the Republic of Uzbekistan. The study’s 15-year duration spans from 2010 until 2024. Box-Jenkin’s methodology, known as Auto Regressive Integrated Moving Average (ARIMA) approach was used in this study on a number of specific Uzbek agricultural commodities. According to the methodology requirement for large amount of data points, 10 species of agricultural products were chosen. All the essential steps of the model were utilized methodically for dynamic forecasting 5 periods ahead from 2025onwards. Using various model selection criteria, including Adj R^ 2, the minimum AIC value, and the lowest MAPE values, the study confirmed the accuracy of the model. Barley had the lowest Akaike Information Criterion (AIC) value among these products, while cabbage had the lowest Mean Absolute Percentage Error (MAPE) value.
