時間 ：2017 年 2 月9 日 週四 2:00 – 4:00pm
地點 ：中研院 人社中心 前棟 3 樓 焦點團體室
venue：Focus Group Room, 3F, RCHSS, Academia Sinica
題目 ： 主力農家所得調查中之抽樣設計
speaker ：Chang-Tai Chao (Dept. of Statistics, National Cheng Kung University)
摘要： Due to the enormous cost and effort required for the survey execution and data processing, the Agriculture, Forestry, Fishery and Animal Husbandry Census can only be conducted every five years. Therefore, certain annual agriculture sampling survey is necessary for the realization of the current related industry information, so that proper policy can be formulated timely. The Primary Farm Household Income Survey has been conducted annually for such purpose since 2013. The target population is the collection of the farm households with certain scale, and the sampled population is formed based on the 2010 census data. A stratified random sampling design, in which the optimal stratum boundary and allocation are carried out based on the 2010 census data, is proposed for the purpose to enhance the estimation precision and investigate certain subpopulations of interest. The simulation result indicates that the performance of the proposed stratified sampling is much more advantageous than simple random sampling without replacement as expected. In addition, it is better than the stratified design with optimal allocation and equal-stratum-size stratification boundary under a comparable total sample size. The overview of the Primary Farm Household Income Survey and a detailed description of the sampling design will be given in this talk. On the other hand, another modified two-stage design which is motivated by the execution experience of this survey will be described as well.
主辦單位 ： 中央研究院人社中心調查研究專題中心
Organizer：Center for Survey Research, RCHSS, Academia Sinica