When a model is nonlinear, boostrap testing can be expensive because of the need to perform at least one nonlinear estimation for every bootstrap sample. We show that it may be possible to reduce computational costs by performing only a fixed, small number of Newton steps or artificial...

Bootstrap tests are tests for which the significance level is calculated by some sort of bootstrap procedure, which may be parametric or nonparametric. We provide a theoretical framework in which to study the size distorsions of bootstrap P values. We show that, in many circumstances, the size...

This paper employs response surface regressions based on simulation experiments to calculate asymptotic distribution functions of the tests for cointegration proposed by Johansen. The paper provides accurate tables of asymptotic critical values. A program which can be used to calculate both...

We show that, in many circumstances, the size distortion of a bootstrap P value for a test will be one whole order of magnitude smaller than that of the corresponding asymptotic P value. We also show that, at least in the parametric case, the magnitude of the distorsion will depend on the shape...

The J test for nonnested regression models often works badly as an asypmtotic test, but it generally works very well when bootstrapped. We provide a theroretical analysis of the J test which explains both of these phenomena. We also propose a modified version of the test which works even better...

This paper proposes a new kind of asymmetric GARCh where the conditional variance obeys two different regimes with a smooth transition function. In one formulation variance reacts differently to negative and positive shocks while a second formulation, small and big shocks have separate effects.

We propose a test for spatial correlation in Probit models that is a joint test for exclusion of spatially lagged-dependent variables and absence of spatial-error correlation. We give a maximum-likelihood justification for the test but use a simulations approach rather than relying on its...

In this paper we are interested in inference based on heteroskedasticity consistent covariance matrix estimators, for which the appropriate bootstrap is a version of the wild bootstrap. Simulation results, obtained by a new very efficient methos, show that all wild bootstraps tests exhibit...

This paper considers the Bayesian analysis of threshold regression models. It shows that this analysis can be conducted with simple deterministic numerical integration in one or two dimensions. The shape of the posterior density is greatly determined by the type of threshold and of transition...

Recent results of Cribari-Neto and Zarkos show that bootstrap methods can be successfully used to estimate a heteroskedasticity robust covariance matrix estimator. We show that their bootstrap estimator can be calculated directly, without bootstrapping, and that inference based on it may not be...