基于平衡单水平轮换的连续性抽样估计方法研究

内容提要：针对现存的各种单水平轮换模式和估计方法，本文提出一套统一的平衡单水平轮换模式。在此轮换模式下，引入两类相关关系，运用线性无偏估计方法，并通过使不同类型估计量方差的加权总和最小的方法确定最优系数，从而得到最优线性无偏估计量，不仅能够减少甚至消除估计量偏差的影响，还能使得连续性调查的整体抽样误差最小，适合估计各种类型的估计量。

The Study of Successive Sampling Estimation Methods Based on Balanced One-Level Rotation

Abstract: This paper puts forward a unified balanced one-level rotation scheme according to different one-level rotation schemes and estimation methods. Under this balanced rotation scheme, two types of correlation and the method of linear unbiased estimation are considered. Through minimizing the weighted sum of variances, we can get one set of the best coefficients and obtain best linear unbiased estimators which not only reduce, even eliminate the bias of estimators, but also make the whole of sampling error from successive survey reach the least and are fitted to estimate different types of estimators.