Bootstrap Methods for Non-nested Hypothesis Tests
Mebane, Walter R.
Cox (1961; 1962) proposed a fairly general method that can be used to
construct powerful tests of alternative hypotheses from separate statistical
families. We prove that non-parametric bootstrap methods can produce
consistent and second-order correct approximations to the distribution of the
Cox statistic for non-nested LISREL-style covariance structure models. We use
the method to investigate a question about the specification of a LISREL model
used by Kinder, Adams and Gronke (1989). In a second application---a pair of
non-nested endogenous switching regression models with tobit-style censoring,
applied to real data---we illustrate how bootstrap calibration can be used to
correct the size of the test when the test distribution is being estimated by
Monte Carlo simulation due to concern about nonregularity.
Endogenous Switching Regression
Document ID Number