FORECASTING Of UNEMPLOYMENT IN UKRAINE ON THE BASIS OF THE BOX–JENKINS METHOD

Abstract

The article analyzes the main modern tendency and approaches to the economic-mathematical modeling of unemployment as a complex socio-economic phenomenon. The analysis of modern domestic and foreign research in this field shows that this task is being given sufficient attention, while the most popular approach for forecasting of unemployment remains an econometric approach, based on a quantitative description of causal relationships between unemployment rates and other factors. At the same time, there are practically no studies based on the use of the theory and tools of time series, which determines the relevance of this direction in the study of unemployment in Ukraine.
In the process of the study, a statistical analysis of the time series of the number of unemployed people in Ukraine was made, which found that the time series of annual unemployment data in Ukraine for the period from 2000 to 2017 may be represented by an additive model of a non-stationary time series that includes trend and random components. Proceeding from this, the method of Box-Jenkins was chosen for short-term forecasts for the number of unemployed in Ukraine. The choice of this method was also based on its universality, since it does not provide any particular single time-series structure on the basis of which the forecast is made.
Using the Box-Jenkins methodology, several alternative specifications of predictive ARIMA models are considered. Estimation of the parameters of these models, their statistical analysis and the verification of adequacy are made with the help of the econometric package EViews 8. Among all possible specifications for forecasting the number of unemployed, the ARIMA model (1,2,0) has been finally selected as the one that satisfies all the criteria for adequacy and efficiency. Based on this model, the forecast of the number of unemployed in Ukraine in 2018 is constructed, which correlates well enough in quantitative and qualitative terms with the general tendency of unemployment growth in Ukraine over the last 3 years.