Hsein Kew

Heteroskedasticity Robust Testing for a Fractional Unit Root

This paper shows how fractional unit root tests originally derived under stationarity can be made robust to heteroskedasticity. This is done by using existing tests nested in a regression framework and then implementing these tests using Whites heteroskedasticity consistent standard errors. We show this approach is effective both asymptotically and in finite samples. We also provide some evidence on the asymptotic local power of different implementations of the tests, both under homoskedasticity and heteroskedasticity.

Confidence Intervals For Policy Reforms In Behavioural Tax Microsimulation Modelling

This paper addresses the need for a measure of the uncertainty that is associated with the results calculated using tax policy behavioural microsimulation models. Deriving the analytical measure would be extremely complicated. Therefore, a simulated approach is proposed, which approximates the sampling distribution of aggregate measures based on the sampling distribution of the estimated labour supply parameters. This approach, which is very computer intensive, is compared with a more time-efficient approach where the functional form of the sampling distribution is assumed to be normal. The results show that in many instances the results from the two approaches are quite similar. The exception is when aggregate measures for minor types of payments, involving relatively small groups of the population, are examined.

ADAPTIVE LONG MEMORY TESTING UNDER HETEROSKEDASTICITY

This paper considers adaptive hypothesis testing for the fractional differencing parameter in a parametric ARFIMA model with unconditional heteroskedasticity of unknown form. A weighted score test based on a nonparametric variance estimator is proposed and shown to be asymptotically equivalent, under the null and local alternatives, to the Neyman-Rao effective score test constructed under Gaussianity and known variance process. The proposed test is therefore asymptotically efficient under Gaussianity. The finite sample properties of the test are investigated in a Monte Carlo experiment and shown to provide potentially large power gains over the usual unweighted long memory test.