The influence of the input parameters variation of the non-seasonal ARIMAX model on the accuracy of meteorological parameters forecasting

Abstract

ARIMA (autoregressive integrated moving-average), one of the most popular models for time-series modeling, is recently frequently used for the needs of events forecasting and prediction in business, medicine, meteorology and engineering domains. In this paper, we present the results of testing the non-seasonal ARIMAX model for short-time forecasting of two meteorological parameters: wind speed and ambient temperature. For the needs of the forecasting accuracy comparison, we propose the use of the MAE (Mean Absolute Error). Various kinds of script files were made using R-Studio, the development environment for testing purposes.