Bootstrap Testing for Fractional Integration

Asymptotic tests for fractional integration, such as the Geweke-Porter-Hudak test, the modified rescaled range test and Lagrange multiplier type tests, exhibit size-distortions in small-samples. This paper investigates a parametric bootstrap testing procedure, for size-correction, by means of a computer simulation study. The bootstrap provides a practical method to eliminate size-distortions in the case of an asymptotic pivotal statistic while the power, in general,is close to the corresponding size-adjusted asymptotic test. The results are very encouraging and suggest that a bootstrap testing procedure does correct for size-distortions.