Country risk forecasting based on EMD and ELM: evidence from BRICS countries

Country risk is an important factor influencing the international investments and transactions. Forecasting country risks of host countries are crucial for investors to make investment strategies and decisions. Considering the complexity and nonlinearity of country risk, this paper proposes a hybrid forecasting model based on empirical mode decomposition (EMD) and extreme learning machine (ELM). Firstly, the original data is decomposed into several different frequency components using EMD. Then, ELM is used to predict the components of different scales respectively, and finally, final country risk forecasting values are integrated. Taking BRICS countries as sample, empirical results show that the EMD-ELM approach performs better than the single prediction models such as ARIMA, SVR and ELM.