排序方式: 共有30条查询结果,搜索用时 15 毫秒
1.
基于错误概率达到最小的原则,该文在罐子模型序贯试验中构造了一种渐近最优设计。在这种设计下,不仅能使病员以较多机会分配到较好的处理,而且能使估计量的统计效率在一定意义下达到最优。 相似文献
2.
白志东 《数学年刊A辑(中文版)》1983,(6)
考虑线性模型y_j=x′β+e_j,j=1,2,…,n,假定误差序列{e_j}为ⅱd,随机变量序列,满足Ee_1=0,00,n→∞; (ⅱ)对任何ε>0,n→∞; (ⅲ) 这个结果与{Y_j}为独立同分布场合完全一致。 相似文献
3.
4.
5.
在通常的线性模型y_i=x′_iβ+e_i(i=1,…,n,…)中,设σ~2=Var(e_i),由前n次观测值y_1,…,y_n,可得基于残差平方和的σ~2的估计(?)_n~2,本文证明了:若随机误差序列独立同分布,则对某个t≥1,E|e_1~2|~t<∞的充要条件为,对任给的ε>0,这样,对于(?)_n~2-σ~2的收敛速度,得到与同分布独立和情形同样优良的结果。 相似文献
6.
In P.L.Hsu(1941),the proof of the basic Lemma 3 is basedon Lemma 1 which is wrong.The aim of this note is to correct the proof ofLemma 3,consequently,to ensure the main theorem in P.L.Hsu(1941). 相似文献
7.
8.
9.
In this paper we have investigated and proved various Lebesgue properties of the n-dimensional quasi-distributions. 2-dimensional quasi-distributions have been classified into 5 types. It is proved among others that every 2-dimensional quasi-distribution can be expressed uniquely as a sum of 5 functions belonging to different types. Some necessary and sufficient conditions for functions belonging to each given type have been analysed in detail. 相似文献
10.
The best possible rate of convergence of the distributions of error variance estimates in linear models, based on the residual sum of squares, is obtained under weakest possible conditions. 相似文献