Markov regime switching and unit-root tests

We investigate the power and size performance of unit-root tests when the data undergo Markov regime switching. All tests, including those robust to a single break in trend growth rate, have low power against a process with a Markov-switching trend. Under the null hypothesis, we nd that previously documented size distortions in Dickey–Fuller-type tests caused by a single break in trend growth rate or variance do not generalize to most parameterizations of Markov switching in trend or variance. However, Markov switching in variance can lead to overrejection in tests allowing for a single break in the level of trend. KEY WORDS: Business-cycle asymmetry; Deterministic trend; Heteroscedasticity; Stochastic trend. For the past 20 years the question of whether various eco-nomic time series have a unit root or are (trend) station-ary has generated much research. Using standard tests, many researchers are unable to reject the unit-root null hypothesis for macroeconomic and nancial time series such as gross domestic product (GDP), interest rates, and exchange rates (Nelson and Plosser 1982). Perron (1989) argued that th