Modeling of Cyclical Economic Processes on the Basis of Spectral Analysis of Time Series

The final qualifying work is devoted to the problem of studying the cyclical nature of the developing economy of Russia. The key objective of the work was to identify elements of cyclicality in development and to model the dynamics of macroeconomic indicators of the Russian economy. Determination of the period of the economic cycle and modeling of output volumes were carried out using two methods - spectral (frequency) analysis and econometric tools. The first part of the paper provides a brief overview of relevant research on economic cyclicality using spectral analysis. The second part of the work is devoted to the theoretical foundations of time-frequency analysis. The practical part of the work consisted in analyzing and modeling statistical data on the economic indicators of Russia: real GDP, CPI, M2 money supply, CPI, unemployment rate, exchange rates (euro and US dollar). Based on the wavelet transformations, it was illustrated that the periods of fluctuations of indicators in the conditions of the modern economic environment are gradually decreasing, which significantly complicates the task of modeling and forecasting cyclicality. It was concluded that the regularity of fluctuations of key indicators is described by a frequency of 5-6 years and the relationship between the cyclicality of output and CPI, M2, unemployment and exchange rates is quite high. This paper shows that wavelet analysis in economic research is an effective tool for improving research methodology, modeling and forecasting economic phenomena. The study was conducted using MS Excel software, Eviews econometric modeling package, Python programming language and Anaconda development environment.