摘要
文献中回归参数线性假设的F-检验统计量主要包括基于广义最小二乘估计F- 统计量F(θ),基于最小二乘估计的F-统计量FLSE以及Wu C．F．J．等于1988年提出的调整的F-统计量FA(θ)．其中后两者因形式简单而常常被广泛采用．本文主要研究了FA(θ)和FLSE的最优性,并分别获得了FA(θ)=F(θ)和ELSE=F(θ)的充要条件．最后,我们将所得的结果应用到医药领域的两类重要模型．
In the literature, F-statistics used in testing linear restrictions mainly include F（θ） based on the general least squares estimate, FLSE based on the least squares estimate, and the adjusted F-statistic FA（θ） given by C. F. J. Wu, et al. in 1988. Due to their simple forms, the latter two are more commonly employed in practice. In this paper, we mainly consider the optimality of FA（θ） and FLSE, and obtain the necessary and sufficient conditions for FA（θ） = F（θ） and FLSE = F（θ）, respectively. Lastly, we apply the results obtained to two important models in the field of pharmacology.
出处
《数学学报（中文版）》
SCIE
CSCD
北大核心
2006年第3期595-604,共10页
Acta Mathematica Sinica：Chinese Series
基金
国家自然科学基金资助项目(10271010)北京市自然科学基金资助项目(1032001)
