CAJ | 학술논문

一、引言　1997年,美国农业部国家农业统计署（NASS）专家Jeffrey T.Bailey和Phillip Kott提出MPPS抽样方法,用于解决多目标抽样问题,即：在有限的资源条件下,为得到多个目标变量的值,如果按单个目标进行抽样,对数据和信息的需求使得调查次数和（或）调查内容增加,从而无法体现抽样调查高效与灵活的优点;MPPS方法采用了多变量与抽样单元规模成比例的概率抽样,实现了多目标抽样.
일、인언　1997년,미국농업부국가농업통계서（NASS）전가Jeffrey T.Bailey화Phillip Kott제출MPPS추양방법,용우해결다목표추양문제,즉：재유한적자원조건하,위득도다개목표변량적치,여과안단개목표진행추양,대수거화신식적수구사득조사차수화（혹）조사내용증가,종이무법체현추양조사고효여령활적우점;MPPS방법채용료다변량여추양단원규모성비례적개솔추양,실현료다목표추양.
MPPS sampling design is a kind of new method that was deVeloped by National Agricultural Statistics Service of Agricultural Department in the U. S. A. at the end of last century. Some ideas about the improvement of the variance estimator of MPPS design are proposed in this paper. The model has been simulated with the data of the First Agricultural Census in China.