This book highlights the importance of studying similarity of business cycles across countries and answers the theoretical question about the behavior of fluctuations in economic activity over different phases of business cycles. This is done by analyzing cross-country data that provides sufficient empirical justifications on the behavior of economic activity to conclude that business cyles are alike. Further, the book maintains, from the recent empirical research, that business cycles fluctuations are asymmetric.
For empirical validation of the hypothesis that business cycles are asymmetric at least in the group of seven highly developed industrialized (G7) countries, real GDP growth rates from these countries are analyzed using nonlinear time series and switching time series models as well as in-sample and jackknife out-of-sample forecasts from neural networks. While importance and application of nonlinear and switching time series models are employed for testing possible existence of business cycle asymmetries in all the series after taking into account long memory, conditional heteroskedasticity, and time varying volatility in the series, usefulness of nonparametric techniques such as artificial neural networks forecasts are discussed and empirically tested to conclude that forecasts from neural networks are superior to the selected time series models. Additionally, the book presents a robust evidence of business cycle asymmetries in G7 countries, which is indeed, the answer to the basic research question on the behavior of economic fluctuation over the business cycles. The book compares spill over and contagion effects due to business cycle fluctuations within the countries studied. In addition, having known the type of business cycle asymmetries,policy makers, empirical researchers, and forecasters would be able to employ appropriate forecasting models for forecasting impact of monetary policy or any other shock on the economies of these countries.