WILD BOOTSTRAPPING IN FINITE POPULATIONS WITH AUXILIARY INFORMATION

Consider a finite population rt, which can be viewed as a realization of a super-population model. A simple ratio model (linear regression, without intercept) with heteroscedastic errors is supposed to have gen erated u. A random sample is drawn without replacement from u, In this set-up a two-stage wild bootstrap resampling scheme as well as severa l other useful forms of bootstrapping in finite populations will be co nsidered. Some asymptotic results for various bootstrap approximations for normalized and Studentized versions of the well-known ratio and r egression estimator are given. Bootstrap based confidence intervals fo r the population total and for the regression parameter of the underly ing ratio model are also discussed.