ESTIMATION WITH DIFFERENT STRATIFICATION AT 2 OCCASIONS

Since forest surveys may use different strata over time, permanent plo ts may be remeasured and new plots may be selected with different prob abilities. The inclusion probabilities and joint probabilities of sele cting both remeasured and new plots, considering both occasions, are d erived. The Horvitz-Thompson estimator using two approximations to act ual probabilities of selection and a stratified sampling estimator ign oring initial probabilities of selection are compared for bias and eff iciency. The Yates-Grundy, probability proportional to size with repla cement, jackknife, and stratified sampling variance estimators are eva luated in terms of bias in estimating the true variance of the estimat ors. Three populations are constructed with different rates of movemen t between strata from time 1 to time 2 to evaluate the estimators. The stratified sampling estimator performed poorly. The jackknife varianc e estimator is recommended.