Analysis and Modeling of Cyclical Processes in the Economy using Wavelet-transforms

Final qualifying work is devoted to the problem of studying the cyclical nature of the economic development of various countries. In particular, attention was paid to the advanced, most dynamically developing economies of the world, such as the USA, Great Britain, Germany, Canada and Japan. The key objective of the work was to identify trends in economic development and modeling of economic cycles based on the analysis of macroeconomic indicators. Determining the nature of fluctuations in gross domestic product and modeling the economic cycle was carried out using two methods such as spectral (frequency) analysis and econometric tools. The first part of the paper presents a brief overview of existing theories and models of cyclicity. The second part of the work is devoted to the theoretical aspects of analysis and modeling of time series. The third chapter presents the key theoretical aspects of spectral analysis. The practical part of the work was to analyze and model statistical data on the GDP of five countries. Based on the wavelet transformations, it was illustrated that the periods of fluctuations in output are gradually reduced in the current economic conditions, which significantly complicates the task of modeling and predicting cyclicity. Using wavelet analysis, it was proved that it is not always possible to trace clear cyclical components in the dynamics of indicators of economic development, which is consistent with current concepts of cycles as more stochastic than deterministic processes. This paper shows that wavelet analysis in economic research is an effective tool for improving the methodology of research, modeling and forecasting of economic phenomena. The study was conducted using MS Excel software, Eviews econometric modeling package, Python programming language and Anaconda development environment.