|
Abstract:
|
A new biased bootstrap methodology for moment condition models has been developed. This biased bootstrap is a form of weighted bootstrap with the weights chosen to satisfy the constraints imposed by the model. First, a least favorable parametric family of weighted empirical distributions associated with the model is constructed. By resampling from this family, we "mimic" the parametric bootstrap for semiparametric models. Using the recycling algorithm, an iterated biased bootstrap is proposed, which is computationally feasible and has a higher degree of accuracy. An extension of this methodology for dependent observations is also presented. Some consistency results, applications to bias reduction and optimal block size selection are provided.
|
