Testing for regime shifts in short-sample African macroeconomic data: a survey of some Monte Carlo evidence

This paper provides a brief survey of issues associated with testing or regime shifts in the short-sample macroeconomic time series data found for African economies. Using Monte Carlo method the paper suggests that the power of standard tests on model stability is extremely low applied to the data sets and regime shifts typically encountered in the macroeconomics of Africa. As a result, standard application of these tests would indicate a false degree of model stability in the face of (true) regime shifts.