统计研究

• 论文 • 上一篇    下一篇

中国广义回归抽样估计系统的构建及应用

陈光慧   

  • 出版日期:2015-07-15 发布日期:2015-07-17

The Construction and Application of the Generalized Regression Estimation System in China

Chen Guanghui   

  • Online:2015-07-15 Published:2015-07-17

摘要: 在抽样理论和应用研究方面,中国一直比较重视抽样方案设计,而忽视抽样估计方法研究。本文在系统总结加拿大等西方国家成功经验的基础上,引入并改进了一套广义回归估计系统,应用在复杂的连续多阶抽样调查中。本文以各类常见的抽样设计为基础,通过模型组和模型水平将现有的超总体模型进行扩展,建立各种类型的回归模型进行模型辅助的广义回归估计,最终形成一套广义回归估计系统,为中国抽样估计的应用研究奠定理论基础。最后,本文以中国农产量的连续多阶抽样调查为例,给出了具体的回归估计程序,从而验证这套系统的实践性和应用价值。

关键词: 超总体模型, 广义回归估计量, 多阶抽样, 连续性抽样, 农产量调查

Abstract: In the research of sampling theory and application, China has been more emphasis on the sampling scheme design, and ignored the study about sampling estimation method. In this paper, on the basis of successful experience systematically from Canada and other western countries, a set of generalized regression estimation system is introduced and improved, and applied to the complex successive multistage sampling survey, in order to lay a theoretical foundation for the application of China’s sampling estimation. Taking the China’s crops multistage sampling survey as an example, we give a set of specific regression estimation procedures in order to verify this system's practicality and application value.

Key words: Superpopulation Model, Generalized Regression Estimator, Multistage Sampling, Successive Sampling, Crops Survey