power problems, whereas the remaining two areadequate and in fact equivalent. The equivalence between the two valid …

method can have higher power than a test for a univariate time series, especially for short time series. Therefore our test …

,hereby investigating the robustness of cointegration methods. Finally, weillustrate how to obtain local power functions of cointegration …

It is generally believed that for the power of unit root tests, only the time span and not the observation frequency … methods will yield power gains in the presence of fat tails and persistent volatility clustering, and the strength of these … features (and hence the power gains) increases with the observation frequency. This is illustrated using both Monte Carlo …

investigating the robustness of cointegration methods. Finally, we illustrate how to obtain local power functions of cointegration …

power problems, whereas the remaining two are adequate and in fact equivalent. The equivalence between the two valid …