Abstract
This article considers standard error estimation of the method of simulated moment estimator for generalized linear mixed models. In literature, parametric bootstrap is used to estimate the covariance matrix, in which we use the estimator to generate simulated moments. To avoid the bias introduced by estimating the parameter and to deal with the correlated observations, we propose a two-stage block nonparametric bootstrap to estimate the standard errors. It is shown from simulation study that the proposed method performs well.