方差分量模型中回归系数的线性估计的可容许性的若干结果

考虑方差分量模型EY=Xβ,VAB(Y)=sum from i=1 to m θ_iV_i,其中n×p矩阵X和非负定矩阵V_i(i=1,2,…,m)都是已知的,β∈R~p,θ_i≥0或θ_i>0(i=1,2,…,m)均为参数.设Sβ是线性可估的。在本文中,我们分别获得了在二次损失和矩阵损失下,关于Sβ的线性估计在线性估计类中可容许的若干结果,并在正态假设下,我们也讨论了线性估计在一切估计类中的可容许性。     

线性约束下部分线性模型中估计的渐近性质

考虑回归模型yi=x′iβ+ g(ti) + ei, 0 ≤i ≤nr=Rβ其中(xi,ti)是固定非随机设计点列,xi=(xi1,…,xip)′,β=(β1,…,βp)′(p 1) ,g是定义在[0 ,1]上的未知函数,β是未知待估参数,0≤ ti≤1i,ei 是i.i.d随机误差,且Eei=0 ,Ee2i=σ2 <∞.r是一个J维向量,R是一个J* p列满秩矩阵,基于g的估计取一个非参数权估计,本文讨论了在线性约束下β的最小二乘估计的相合性及渐近正态性.     

矩阵损失下约束线性模型中回归系数的线性可容许估计

考虑约束线性模型Mr={Y,Xβ,σ2V|Rβ=r}其中x列满秩,V为正定矩阵.在二次损失下,Baksalary. J. K和Markiewicz,A得到了回归系教β的线性估计在非齐次线性估计类中可容许的充分必要条件,利用吴启光在无约束线性模型关于回归系数线性可容许估计的结果,对约束线性模型Mr我们得到结果如下在矩阵损失下回归系数β的线性估计AY+g在非齐次线性估计类中可容许当且仅当[i]XAV对称;[ii]R(A) R(U) [iii]AXU=U,g=(AX-I)R+r或AXU≠U时,有r(AX) (-∞,0)∪(1,+∞).其中R(U)=N(R),U为列正交矩阵.     

加权平方和损失下线性估计的可容许性

1980年,Berger讨论了Γ分布尺度参数的通常估计的容许性向题.本文在此基础上讨论Γ分布尺度参数的线性估计的容许性问题,即 例1 设X_1和X_2相互独立,X_1～Γ(α_1,β_1~(-1)),X_2～Γ(α_2,β_2~(-1))α_1和α_2是已知的正常数,β=(β_1,β_2)′∈R~ ×R~ 是未知的参数.取β的估计为线性估计     

带有异方差的部分线性回归模型的B样条估计

考虑一带有异方差的固定设计部分线性回归模型yij=X'ijβ+g(tij)+εij,i=1,2…,k:j=1,2,…,ni,和sum from i=1 to kni=n,其中yij为响应变量,β=(β1,…,βp)’是未知的参数向量,g(·)是未知的函数,Xij=(Xij1,…,Xijp)’和tij∈[0,1]为已知的非随机设计点,εij为均值0,方差是σi2的随机误差,其中σi2可能不同.通过B样条级数近似非参数分量,构造了参数分量β的一个半参数广义最小二乘估计.在一些矩条件下,导出了此半参数广义最小二乘估计的渐近分布,大多数在实际中遇到的误差分布都满足这些矩条件.另外,也构造了半参数广义最小二乘估计的渐近协方差矩阵的一个相合估计,还讨论了非参数分量的B样条估计.所有这些大样本性质都是在k趋于无穷大,ni有限时导出的.这些结果能被用来做渐近有效的统计推导.     

线性回归模型参数的联立经验Bayes估计的收敛速度

本文考虑了线性模型中回归系数β=(β₁,…,β_p)′和误差方差σ²的联立经验Bayes(EB)估计.在二次损失下,利用密度函数及其编导数的核估计构造出参数θ=(β₁,…,β_p,σ²)的联立EB估计,在一定条件下证明了θ的联立EB估计的收敛速度任意接近于1.最后、给出了一个实例.     

一、二阶矩同时估计的可容许性

考虑模型Y=(y_1,…,y_n)′=(β,…,β)′+(ε_1,…,ε_n)′=1β+ε.(1.1)此处1=(1,…,1)′;ε_1,…,ε_n 相互独立,E(ε_i)=0,E(ε_i~2)=σ~2,E(ε_i~3)=0,E(ε_i~4)=3σ~4,i=1,…,n;-∞<β<∞,0<σ<∞.鉴于 β 的最重要的估计量是观察值 Y 的线性函数,σ~2和 β~2+σ~2的最重要的估计量是 Y 的非负定二次型,在考虑 β 的估计时,首先把注意力集中在 Y 的线性函数上;在考虑σ~2或 β~2+σ~2的估计时,首先考虑 Y 的非负定二次型.参考文献[1]在一般线性模型和二次损失下,给出了回归系数的可估线性函数的估计在线性估计类中是可容许的充要条件.参考文献[2]和[3]在模型(1.1)和平方损失下给出了 σ~2的估计在非负定二次型估计类中是可容许的充要条件;而在一般线性模型和平方损失下,给出了 σ~2的估计在非负定二次型估计类中是可容许的必要条件和充分条件,给出了相当大的一类可容许估计;此外,给     

带有异方差的部分线性回归模型的B样条估计

考虑一带有异方差的固定设计部分线性回归模型yij=x'ijβ+g(tij)+εij,i=1,2,…,k,j=1,2,…,ni,和∑ni=n,其中yij为响应变量,β=(β1,…,βp)'是未知的参数向量,g(.)是未知的函数,xij=(xij1,…,xijp)'和tij∈[0,1]为已知的非随机设计点,εij为均值0,方差是σ2i的随机误差,其中σ2i可能不同.通过B样条级数近似非参数分量,构造了参数分量β的一个半参数广义最小二乘估计.在一些矩条件下,导出了此半参数广义最小二乘估计的渐近分布,大多数在实际中遇到的误差分布都满足这些矩条件.另外,也构造了半参数广义最小二乘估计的渐近协方差矩阵的一个相合估计,还讨论了非参数分量的B样条估计.所有这些大样本性质都是在k趋于无穷大,n.有限时导出的.这些结果能被用来做渐近有效的统计推导.     

具有P个方差分量的线性模型的Beyes二次估计及可容许估计的非负性

本文对具有 p 个方差分量的线性模型讨论了方差分量线性函数的 Bayes 不变二次估计问题,给出了 Bayes 不变二次估计(无偏和有偏)的显示表达式,并且证明了它们在各自考虑的类中形成了可容许估计的完全类.在可容许估计的完全类中,还讨论了非负参数函数的非负估计问题,给出了可容许的非负定估计存在的充要条件.     

线性回归模型Liu估计的新研究

考虑线性回归模型Y=Xβ+ε,E()ε=0,Cov()ε=2σI(1),当设计矩阵X的列存在共线性时,最小二乘估计^β=(X′X)-1X′Y的性质变坏,为此给出了有偏估计^(βK,d)=(X′X+K)-1(X′Y+d^β),其中K为对角矩阵,K=diag(k1,…kp),ki≥0,d>0为参数,讨论了这种有偏估计与广义岭估计、Liu估计的比较,并证明了其可容许性估计.     

ADMISSIBLE ESTIMATES IN THE IMPORTANT CLASS OF ESTIMATES OF THE COVARIANCE MATRIX

Let X1,…XN(where N&gt;m)be independent Nm(μ,∑）random vectors,and put X^-=1/N ∑i=1^N Xi and T‘T=A=∑i=1^N(Xi-X^-)(Xi-X^-)‘,where T is upper-triangular with positive diagonal elements.The author considers the problem of estimating ∑,and restricts his attention to the class of estimates D={T‘△^*T+Nb^*X^-X^‘&#183;△^* is any diagonal matrix and b^* is any nonnegative constant}because it has the following attractive features:(a)Its elements are all quadratic forms of the sufficient and complete statistics(X^-,T).(b)It contains all estimates of the form αA+NbX^-X^-‘(α≥0 and b≥0),which construct a complete subclass of the class of nonnegative quadratic estimates D^8={X‘BX:B≥0}(where X=(X1,…,XN)‘)for any strict convex loss function.(c)It contains all invariant estimates under the transformation group of upper-triangular matrices.The author obtains the characteristics for an estimate of the form.T‘△T+NbX^-X^-‘(△=diag{δ1,…，δm}≥0 and b≥0)of ∑ to be admissible in D when the loss function is chosen as tr(∑^-1∑-I)^2,and shows,by an example,that αA+NX^-X^-‘(α≥0 and b≥0)is admissible in D^* can not imply its admissibility in D.     

一类连分数的有理逼近

设f(n)是非负函数,k,b,s_i,t_i(i=1,2,…)是正常数,研究形如[a_0,a_1,a_2…]=[■]_m~∞=0和[■]_n~∞=1的连分数有理逼近的下界．     

关于正态分布的次序统计量的随机序

设$X_1,X_2,\cdots,X_n$和$X^*_1,X^*_2,\cdots,X^*_n$分别服从正态分布$N(\mu_i,\sigma^2)$和$N(\mu^*_i,\sigma^2)$,以$X_{(1)}$,$X^*_{(1)}$分别表示$X_1,\cdots,X_n$和$X^*_1,\cdots,X^*_n$的极小次序统计量,以$X_{(n)}$, $X^*_{(n)}$分别表示$X_1,\cdots,X_n$和$X^*_1,\cdots$,$X^*_n$的极大次序统计量. 我们得到了如下结果:(i)\,如果存在严格单调函数$f$使得$(f(\mu_{1}),\cdots,f(\mu_{n}))\succeq_{\text{m}}$ $(f(\mu^{*}_{1}),\cdots,f(\mu^{*}_{n}))$,且$f'(x)f'(x)\!\geq\!0$, 则$X_{(1)}\!\leq_{\text{st}}\!X^*_{(1)}$;(ii)\,如果存在严格单调函数$f$使得$(f(\mu_{1})$,$\cdots,f(\mu_{n}))\succeq_{\text{m}}(f(\mu^{*}_{1}),\cdots,f(\mu^{*}_{n}))$,且$f'(x)f'(x)\leq 0$, 则$X_{(n)}\geq_{\text{st}}X^*_{(n)}$.(iii)\,设$X_{1},X_{2},\cdots,X_{n}$和\, $X^*_{1},X^*_{2},\cdots,X^*_{n}$分别服从正态分布$N(\mu,\sigma_i^2)$和$N(\mu,\sigma_i^{*2})$,若$({1}/{\sigma_{1}},\cdots,{1}/{\sigma_{n}})\succeq_{\text{m}}({1}/{\sigma^{*}_{1}},\cdots,{1}/{\sigma^{*}_{n}})$,则有$X_{(1)}\leq_{\text{st}}X^*_{(1)}$和$X_{(n)}\geq_{\text{st}}X^*_{(n)}$同时成立.     

拟线性次椭圆方程组在Morrey空间上的部分正则性

证明了拟线性次椭圆方程组-X_α~*(a_(ij)~(αβ)(x,u)X_βu~j)=-X_α~*f_i~α+g_i,i=1,2,…,N,x∈Ω的弱解广义梯度Xu在Morrey空间L_x~(p,λ)(Ω,R~(mN))(p2)上的部分正则性,其中光滑实向量场族X=(X_1,X_2,…,X_m)满足H(o|¨)rmander有限秩条件,X_α~*是X_α的共轭;而且主项系数a_(ij)~(αβ)(x,u)关于x一致VMO(Vanishing Mean Oscillation的缩写,消失平均震荡)间断,且关于u为一致连续.     

On effective σ‐boundedness and σ‐compactness

We prove several dichotomy theorems which extend some known results on σ‐bounded and σ‐compact pointsets. In particular we show that, given a finite number of $\Delta ^{1}_{1}$ equivalence relations $\mathrel {\mathsf {F}}_1,\dots ,\mathrel {\mathsf {F}}_n$, any $\Sigma ^{1}_{1}$ set A of the Baire space either is covered by compact $\Delta ^{1}_{1}$ sets and lightface $\Delta ^{1}_{1}$ equivalence classes of the relations $\mathrel {\mathsf {F}}_i$, or A contains a superperfect subset which is pairwise $\mathrel {\mathsf {F}}_i$‐inequivalent for all i = 1, …, n. Further generalizations to $\Sigma ^{1}_{2}$ sets A are obtained.     

Ultrafilters on the collection of finite subsets of an infinite set

Let $J$ be an infinite set and let $I={\cal P}_{f}( J)$, i.e., $I$ is the collection of all non empty finite subsets of $J$. Let $\beta I$ denote the collection of all ultrafilters on the set $I$. In this paper, we consider $( \beta I,\uplus ),$ the compact (Hausdorff) right topological semigroup that is the {\it Stone-$\check{C}\!\!$ech} $Compactification$ of the semigroup $\left( I,\cup \right)$ equipped with the discrete topology. It is shown that there is an injective map $A\rightarrow \beta _{A}( I) $ of ${\cal P}( J) $ into ${\cal P}( \beta I) $ such that each $\beta _{A}( I) $ is a closed subsemigroup of $ ( \beta I,\uplus ) $, the set $\beta _{J}( I) $ is a closed ideal of $( \beta I,\uplus ) $and the collection $\{ \beta _{A}( I) \mid A\in {\cal P} ( J) \} $ is a partition of $\beta I$. The algebraic structure of $\beta I$ is explored. In particular, it is shown that {\bf (1)} $\beta _{J}\left( I\right) =\overline{K( \beta I) }$, i.e., $\beta _{J}( I) $is the closure of the smallest ideal of $\beta I$, and {\bf (2)} for each non empty $A\subset J$, the set ${\cal V}_{A}=\tbigcup \{ \beta_{B}( I) \mid B\subset A\} $is a closed subsemigroup of $( \beta I,\uplus ) ,$ $\beta _{A}( I) $ is a proper ideal of ${\cal V}_{A},$ and ${\cal V}_{A}$ is the largest subsemigroup of $( \beta I,\uplus ) $ that has $ \beta _{A}( I) $ as an ideal.     

Classification of Solutions to a Critically Nonlinear System of Elliptic Equations on Euclidean Half-Space

For $N\geq 3$ and non-negative real numbers $a_{ij}$ and $b_{ij}$ ($i,j= 1, \cdots, m$), the semi-linear elliptic system\begin{equation*} \begin{cases}\Delta u_i+\prod\limits_{j=1}^m u_j^{a_{ij}}=0,\text{in}\mathbb{R}_+^N,\\dfrac{\partial u_i}{\partial y_N}=c_i\prod\limits_{j=1}^m u_j^{b_{ij}},\text{on} \partial\mathbb{R}_+^N,\end{cases}\qquad i=1,\cdots,m,\end{equation*} % is considered, where $\mathbb{R}_+^N$ is the upper half of $N$-dimensional Euclidean space. Under suitable assumptions on the exponents $a_{ij}$ and $b_{ij}$, a classification theorem for the positive $C^2(\mathbb{R}_+^N)\cap C^1(\overline{R_+^N})$-solutions of this system is proven.     

INJECTIVE MAPS ON PRIMITIVE SEQUENCES OVER Z/（p^e）

Let Z/(pe) be the integer residue ring modulo pe with p an odd prime and integer e ≥ 3. For a sequence (a) over Z/(pe), there is a unique p-adic decomposition (a) = (a)0 (a)1·p … (a)e-1 ·pe-1, where each (a)i can be regarded as a sequence over Z/(p), 0 ≤ i ≤ e - 1. Let f(x) be a primitive polynomial over Z/(pe) and G' (f(x), pe) the set of all primitive sequences generated by f(x) over Z/(pe). For μ(x) ∈ Z/(p)[x] with deg(μ(x)) ≥ 2 and gcd(1 deg(μ(x)),p- 1) = 1,set ψe-1 (x0, x1,…, xe-1) = xe-1·[ μ(xe-2) ηe-3 (x0, x1,…, xe-3)] ηe-2 (x0, x1,…, xe-2),which is a function of e variables over Z/(p). Then the compressing map ψe-1: G'(f(x),pe) → (Z/(p))∞,(a) (→)ψe-1((a)0, (a)1,… ,(a)e-1) is injective. That is, for (a), (b) ∈ G' (f(x), pe), (a) = (b) if and only if ψe - 1 ((a)0, (a)1,… , (a)e - 1) =ψe - 1 ((b)0,(b)1,… ,(b)e-1). As for the case of e = 2, similar result is also given. Furthermore, if functions ψe-1 and ψe-1 over Z/(p) are both of the above form and satisfy ψe-1((a)0,(a)1,… ,(a)e-1) = ψe-1((b)0,(b)1,… ,(b)e-1) for (a),(b) ∈ G'(f(x),pe), the relations between (a) and (b), ψe-1 and ψe-1 are discussed.     

Boundedness of Marcinkiewicz integral related to multilinear commutators with variable kernels on Hardy spaces

Let→b=(b1,b2,…,bm),bi∈∧βi(Rn),1≤I≤m,βi＞0,m∑I=1βi=β,0＜β＜1,μΩ→b(f)(x)=(∫∞0|F→b,t(f)(x)|2dt/t3)1/2,F→b,t(f)(x)=∫|x-y|≤t Ω(x,x-y)/|x-y|n-1 mΠi=1[bi(x)-bi(y)dy.We consider the boundedness of μΩ,→b on Hardy type space Hp→b(Rn).     

Higher Commutators of Pseudo-Differential Operator

In this paper the following result is established: For a_i,f\in \phi(R^K),i=1,\cdots,n and $T(a,f)(x)=w(x,D)()[\prod\limits_{i = 1}^n {{P_{{m_i}}}({a_i},x, \cdot )f( \cdot )} \]$ It holds that $||T(a,f)||_q\leq C||f||_p_0[\prod\limits_{i = 1}^n {||{\nabla ^{{m_i}}}|{|_{{p_i}}}} \]$ where a=(a_1,\cdots,a_n), q^-1=p^-1_0+[\sum\limits_{i = 1}^n {p_i^{ - 1} \in (0,1),\forall i,{p_i} \in (1,\infty )} \] or \forall i,p_i=\infinity,p_0\in (1,\infinity), for an integer m_i\geq 0, $P_m_m(a_i,x,y)=a_i(x)-[\sum\limits_{|\beta | < {m_i}} {\frac{{a_i^{(\beta )}(y)}}{{\beta !}}} {(x - y)^\beta }\]$ w(x,\xi) is a classical symbol of order |m|, m=(m_1,\cdots, m_n), |m|=m_1+\cdots+m_n, m_i are nonnegative integers. Besides, a representation theorem is given. The methods used here closely follow those developed by Coifman, R. and Meyer, Y. in [5] and by Cohen, J. in [3].