Detection of trend-seasonal components in time series of economic processes

Abstract

Purpose: Research of trend-seasonal economic processes on the basis of time series with the subsequent preparation of data for forecasting the dynamics of development using hybrid and nonlinear methods of forecasting. Discussion: In modern conditions of economy forecasting based on time series is often used to minimize risks and losses. One of the most significant factors that must be taken into account at the initial stage of the analysis of time series data is seasonality. In this paper we present an algorithm for identifying the Chetverikov seasonal component; the method was tested on the basis of a reference series and presents a study on the identification of the seasonal component in the time series of prices for Brent crude oil, the dollar-to-ruble exchange rate and the time series of wholesale bottled drinking water in the cities of the Krasnodar region. Results: the Chetverikov method allows us to single out a seasonal component in time series with a clearly existing dependence on the time of year. Talking about seasonality in such time series as Brent prices and the dollar rate is quite difficult, as their value is influenced by geopolitical factors, for example, the arrangements of the OPEC countries. The authors plan to conduct a further study of the time series presented in the paper to predict the dynamics of processes’ development using both hybrid methods and methods of nonlinear dynamics