Heteroskedasticity-robust unit root testing for trending panels

Standard panel unit root tests (PURTs) are not robust to breaks in innovation variances. Consequently, recent papers have proposed PURTs that are pivotal in the presence of volatility shifts. The applicability of these tests, however, has been restricted to cases where the data contains only an intercept, and not a linear trend. This paper proposes a new heteroskedasticity-robust PURT that works well for trending data. Under the null hypothesis, the test statistic has a limiting Gaussian distribution. Simulation results reveal that the test tends to be conservative but shows remarkable power in finite samples.