一种方式分组随机模型中估计方差分量的最小风险设计

§1.引言一种方式分组随机模型:y_(ij)=β α_i ε_(ij),i=1,…,n,j=1,…,m_i,(1.1)其中 ε_(ij)(i=1,…,n,j=1,…,m_i)是相互独立的随机误差,α_i(i=1,…,n)是独立的随机变量.Eα_i=Eε_(ij)=0,varε_(ij)=θ_1>0,varα_i=θ_2≥0,cov(α_i,ε_(ij))=0.β、θ_1、θ_2是未知参数,β∈R~1,(θ_1,θ_2)~T∈Θ(?){θ_1>0,θ_2≥0}.     

对“关于连续介质有限变形问题的几点讨论”的讨论

应变张量ε_(αβ)和θ~(αβ)可以定义在变形前的坐标架中,也可以定义在变形后的坐标架中,相应称为Lagrange应变张量和Euler应变张量,以资区别.当采用直角笛卡尔坐标时,则常称前者为Green-Cauchy张量,后者为Almansi-Hamel张量(有的文献中,也用Love、Segawa、Finger张量等名称不一);它们的分量ε_(αβ)和θ~(αβ)都可以有直观的几何解释(提醒一句:ε_(αβ)和θ~(αβ)不是同一张量的两种分量;ε_(αβ)≠θ_(αβ),ε~(αβ)≠θ~(αβ).     

关于亚纯函数的Hayman方向

定义1 设 f(z)为开平面上ρ(0≤ρ<+∞)级亚纯函数。B:argz=θ_0(0≤θ_0<2π)为原点出发的直线。若对任意正整数 l,任意正数ε,及任意两个有穷复数 a,b(b≠0)(?)(log+{n(r,θ,ε,f=a)+n(r,θ_0,ε,f~(l)=b)})/log     

亚纯函数结合于导数和重值的辐角分布

本文考虑了关于亚纯函数结合其导数涉及重值的辐角分布方面的问题,主要证明了:定理1 设 f(x)是λ级亚纯函数,0<λ<∝,则存在一条由原点出发的半直线 B:arg z=θ_0,(0≤θ_0<2π)使得对于任意正数ε,一切有穷复数 a 与一切有穷非零复数 b 有:(?)(log{n(r,θ_0,ε,f)+n_(k-1)(r,θ_0,ε,f=a)+n_(l-1)(r,θ_0,ε,f~(m)=b)})/log r其中 k,l,m 为正数且满足条件 (m+1)/k+1/l<1.本文还对定理1作了推广。     

Asymtotics of M-estmation in Non-linear Regression

Consider the standard non-linear regression model y_i=g(x_i,θ_0) ε_j,i=1,...,n whereg(x,θ) is a continuous function on a bounded closed region X×θ_0 is the unknown parametervector in R~p,{x_1,x_2,...,x_n} is a deterministic design of experiment and {ε_1,ε_2,...,ε_n} is asequence of independent random variables.This paper establishes the existences of M-estimates andthe asymptotic uniform linearity of M-scores in a family of non-linear regression models when theerrors are independent and identically distributed.This result is then used to obtain the asymptoticdistribution of a class of M-estimators for a large class of non-linear regression models.At the sametime,we point out that Theorem 2 of Wang (1995) (J.of Multivariate Analysis,vol.54,pp.227-238,Corrigenda.vol.55,p.350) is not correct.     

一个基本不等式及相应的奇异方向

本文证明了一个用N(r,1/f)和N(r,1/(F-1))去限制亚纯函数f的特征函数T(r,f)的基本不等式,其中F=f~(k) a_nf~n … a_1f,这里的n和k满足1≤n相似文献   

关于线性型的一个转换定理

<正> 命θ_1,…,θ_n为一组实数,n≥2;1,θ_1,…,θ_n信在有理数域上线性无关.对实数x,命 ‖x‖=min(x-[x],[x]+1-x),x=max(1,|x|).ε表示任意给定的正数,c_1,c_2表示仅依赖于ε,n,θ_1,…,θ_n的正常数.     

Hermite-Fejer Interpolation With Moved End Points

Let Z_n={z_(kn)=cosθ_(kn):θ_(kn)=(2k-1)/(2n)π,k=1,2…,n}be the zeros of T_n(x)=cosnθ(x=cosθ,θ∈[0,π]).For 0≤ε≤1,let α_n=:α_n(ε)=:cos(1-ε)/(2n)π,β_n=:β_n(ε)=:cos(2n-1+ε)/(2n)π=-α_n,X_n~(1)=(Z_n-{z_(1z)})∪{α_n},X_n~(2)=(Zn-{z_(nn)})∪{β_n},X_n~(3)=(Z_n-{z_(1n),z_(nn)})∪{α_n,β_n},Y_n~(1)=Z_n∪{α_n},Y_n~(2)=Z_n∪{β_n},Y_n~(3)=Z_n∪{α_nβ_n}.     

THE STRONG CONSISTENCY OF ERROR PROBABILITY ESTIMATES IN NN DISCRIMINATION

Let(X,θ),(X_1,θ_1),…,(X_n,θ_n)be iid.R~d×{1,2,…,s}-valued random vectors and letL_n be the posterior error probability in NN(nearest neighbor).diserimination.Someknowledge of the unknown value of L_n is of great meaning in many applications.For thisaim,in 1971,T.J.Wagner introduced an estimate of L_n which is defined by_n=1/nI(θ_j≠θ_(nj)),where θ_(nj) is the NN discrimination of θ_j based on the training samples(X_1,θ_1),…,(X_(j-1),θ_(j-1)),(X_(j+1),θ_(j+1)),…,(X_n,θ_n).Then he showed that _nR,where R is the limit ofthe prior error probability.But the problem of“)nR” is still left open since thattime.In this paper,it is shown that for any s>0,there exist two positive constants a andC such that P(丨_n-R丨≥ε)≤Ce~(-an).By this it is clear that _nR.     

ON THE EXPECTED SAMPLE SIZES OF SOME POWER ONE TESTS FOR NORMAL MEAN WITH UNKNOWN VARIANCE

Suppose that x_1, x_2,…axe i. i. d. random variables on a probability space (Ω, ■, P_(θσ)) and x_1 is normally distributed with mean θ and variance σ~2, both of which are unknown. Given θ_0 and 0<α<1, we propose a concrete stopping rule T w. r. t. the {x_n, n≥1} such thatP_(θσ)(T<∞)≤α for all θ≤θ_0, σ>0,P_(θσ)(T<∞)=1 for all θ>θ_0, σ>0,■(θ-θ_0)~2(ln_2 1/θ-θ_0)~(-1)E_(θσ)T=2σ~2P_(θ_0σ)(T=∞),where ln_2 x=In(ln x).