Modelling and Prognosis of the Export of the Republic of Serbia by Using Seasonal Holt-Winters and Arima Method

Abstract Low labour costs as one of the key sources of export stimulation, the competitive advantage of domestic agricultural production and bilateral agreements with partner countries - all promote export as a potentially significant factor of encouragement of economic development of the Republic of Serbia.Taking into account this fact, on the one hand, and balance of payments problems that Serbia has been facing over the years, on the other hand, the subject of this paper is an analysis of trends in the Republic of Serbia export and explanation of variations in the export trends during the period from 2004 to 2014. The aim of the paper is to explore export trends forecast from January to December 2015.The analysis uses Holt-Winters and ARIMA methods for analyszing time series.The paper provides insight into the export trend forecasts for the period of 12 months, and thus confirms the possibility of practical usage of the time series analysis methods in forecasting macroeconomic variables such as export. The used methods identify increase of export during the summer and its decrease after October 2015. The paper establishes the existence of a high degree of congruence between forecasts obtained by using two methods, which confirm a high quality of the elaborated methods in the analysis of exports.