Testing spatial randomness based on empirical distribution function: A study on lattice data

In this article, we extended the empirical distribution function based test statistic _Ik$I_k$ of Skaug and Tjostheim [1993. Nonparametric test of serial independence based on the empirical distribution function. Biometrika 80, 591–602] in the time series setting to _Dn$D_n$ for spatial lattice data and derived the asymptotic distribution of the proposed test statistic _Dn$D_n$ under the null hypothesis of spatial independence. The size and power of the proposed test statistic under conditional autoregressive model (CAR) were simulated. We applied _Dn$D_n$, Moran's I and Geary's c   to the transformed and well-studied sudden infant death syndrome data from North Carolina and found that _Dn$D_n$ produced a much smaller p$p$-value in testing spatial independence.     

Truncated linear estimation of a bounded multivariate normal mean

We consider the problem of estimating the mean θ$\theta$ of an _Np(θ,_Ip)${\mathrm{N}}_{\mathrm{p}}(\theta ,I_p)$ distribution with squared error loss ∥δ−θ∥²$\parallel \delta -\theta {\parallel }^2$ and under the constraint ∥θ∥≤m$\parallel \theta \parallel \le m$, for some constant m>0$m>0$. Using Stein's identity to obtain unbiased estimates of risk, Karlin's sign change arguments, and conditional risk analysis, we compare the risk performance of truncated linear estimators with that of the maximum likelihood estimator _δmle${\delta }_{\mathrm{mle}}$. We obtain for fixed (m,p)$(m,p)$ sufficient conditions for dominance. An asymptotic framework is developed, where we demonstrate that the truncated linear minimax estimator dominates _δmle${\delta }_{\mathrm{mle}}$, and where we obtain simple and accurate measures of relative improvement in risk. Numerical evaluations illustrate the effectiveness of the asymptotic framework for approximating the risks for moderate or large values of p.     

An identity for the Fisher information and Mahalanobis distance

Consider a mixture problem consisting of k classes. Suppose we observe an s-dimensional random vector X   whose distribution is specified by the relations P(X∈A|Y=i)=_Pi(A)$P(X\in A|Y=i)=P_i(A)$, where Y   is an unobserved class identifier defined on {1,…,k}$\{1,\dots ,k\}$, having distribution P(Y=i)=_pi$P(Y=i)=p_i$. Assuming the distributions _Pi$P_i$ having a common covariance matrix, elegant identities are presented that connect the matrix of Fisher information in Y   on the parameters _p1,…,_pk$p_1,\dots ,p_k$, the matrix of linear information in X, and the Mahalanobis distances between the pairs of P  's. Since the parameters are not free, the information matrices are singular and the technique of generalized inverses is used. A matrix extension of the Mahalanobis distance and its invariant forms are introduced that are of interest in their own right. In terms of parameter estimation, the results provide an independent of the parameter upper bound for the loss of accuracy by esimating _p1,…,_pk$p_1,\dots ,p_k$ from a sample of X^′$X^{\prime }$s, as compared with the ideal estimator based on a random sample of Y^′$Y^{\prime }$s.     

Simultaneous estimation of Poisson means of the selected subset

Let _π1,_π2,…,_πp${\pi }_1,{\pi }_2,\dots ,{\pi }_p$ be p   independent Poisson populations with means _λ1,…,_λp${\lambda }_1,\dots ,{\lambda }_p$, respectively. Let {X₁,…,X_p} denote the set of observations, where X_i is from _πi${\pi }_i$. Suppose a subset of populations is selected using Gupta and Huang's (1975) selection rule which selects _πi${\pi }_i$ if and only if _Xi+1?_cX(1)$X_i+1?{\mathit{cX}}_{(1)}$, where X₍₁₎=max{X₁,…,X_p}, and 0<c<1$0<c<1$. In this paper, the simultaneous estimation of the Poisson means associated with the selected populations is considered for the k-normalized squared error loss function. It is shown that the natural estimator is positively biased. Also, a class of estimators that are better than the natural estimator is obtained by solving certain difference inequalities over the sample space. A class of estimators which dominate the UMVUE is also obtained. Monte carlo simulations are used to assess the percentage improvements and an application to a real-life example is also discussed.     

Empirical Bayes vs. fully Bayes variable selection

For the problem of variable selection for the normal linear model, fixed penalty selection criteria such as AIC, _Cp${\mathrm{C}}_p$, BIC and RIC correspond to the posterior modes of a hierarchical Bayes model for various fixed hyperparameter settings. Adaptive selection criteria obtained by empirical Bayes estimation of the hyperparameters have been shown by George and Foster [2000. Calibration and Empirical Bayes variable selection. Biometrika 87(4), 731–747] to improve on these fixed selection criteria. In this paper, we study the potential of alternative fully Bayes methods, which instead margin out the hyperparameters with respect to prior distributions. Several structured prior formulations are considered for which fully Bayes selection and estimation methods are obtained. Analytical and simulation comparisons with empirical Bayes counterparts are studied.     

The q-cluster distribution

The paper studies the three-parameter generalization of the logarithmic distribution that is obtained as the cluster distribution for the generalized Euler distribution. The diagnostic statistic, R(x)=_xpx/[(x-1)_px-1]$R(x)={\mathit{xp}}_x/[(x-1)p_{x-1}]$, is constant for the logarithmic distribution. For the new distribution it can decrease, stay constant, or increase as x increases. The relative values of the extra parameters determine the flatness/hollowness of the distribution and its tail behaviour. Kemp's q-logarithmic distribution and the Euler cluster distribution are special cases. Fitted data sets illustrate the properties of the distribution and its limiting forms.     

Moderate deviations for LS estimator in simple linear EV regression model

In this paper, we consider the following simple linear Errors-in-Variables (EV) regression model _ηi=θ+β_xi+_?i${\eta }_i=\theta +\beta x_i+{?}_i$, _ξi=_xi+_δi${\xi }_i=x_i+{\delta }_i$, 1?i?n$1?i?n$. The moderate deviation principle for the least squares (LS) estimators of the unknown parameters θ$\theta$, β$\beta$ in the model are obtained.     

Optimal designs for threshold-determining limiting dilution assays

In common with other non-linear models, the optimal design for a limiting dilution assay (LDA) depends on the value of the unknown parameter, θ$\theta$, in the model. Consequently optimal designs cannot be specified unless some assumptions are made about the possible values of θ$\theta$. If a prior distribution can be specified then a Bayesian approach can be adopted. A proper specification of the Bayesian approach requires the aim of the experiment to be described and quantified through an appropriate utility function. This paper addresses the problem of finding optimal designs for LDAs when the aim is to determine whether θ$\theta$ is above or below a specified threshold, _θ0${\theta }_0$.     

Double shrink methodologies to determine the sample size via covariance structures

We consider fixed-size estimation for a linear function of mean vectors from _πi:_Np(_μi,_Σi)${\pi }_i:{\mathrm{N}}_p({\mathit{\mu }}_i,{\mathbf{\Sigma }}_i)$, i=1,…,k$i=1,\dots ,k$, when every _Σi${\mathbf{\Sigma }}_i$ has some structure. The goal of inference is to construct a fixed-span confidence region with required accuracy. We find a sample size for each _πi${\pi }_i$ with the help of the ‘double shrink methodology’, that is introduced by this paper, via covariance structures of _Σi${\mathbf{\Sigma }}_i$, i=1,…,k$i=1,\dots ,k$. We estimate the sample size in a two-stage sampling and give a fixed-span confidence region that has the coverage probability approximately second-order consistent with the required accuracy. Some simulations are carried out to see moderate sample size performances of the proposed methodologies.     

Generalizing Clatworthy group divisible designs

In this paper a neat construction is provided for three new families of group divisible designs that generalize some designs from Clatworthy's table of the only 11 designs with two associate classes that have block size four, three groups, and replication numbers at most 10. In each case (namely, _λ1=4${\lambda }_1=4$ and _λ2=5${\lambda }_2=5$, _λ1=4${\lambda }_1=4$ and _λ2=2${\lambda }_2=2$, and _λ1=8${\lambda }_1=8$ and _λ2=4${\lambda }_2=4$), we have proved that the necessary conditions found are also sufficient for the existence of such GDD's with block size four and three groups, with one possible exception.     

Design selection criteria for discrimination/estimation for nested models and a binomial response

The aim of an experiment is often to enable discrimination between competing forms for a response model. We investigate the selection of a continuous design for a non-sequential strategy when there are two competing generalized linear models for a binomial response, with a common link function and the linear predictor of one model nested within that of the other. A new criterion, _TE$T_E$-optimality, is defined, based on the difference in the deviances from the two models, and comparisons are made with T$T$-, _Ds$D_s$- and D$D$-optimality. Issues are raised through the study of two examples in which designs are assessed using simulation studies of the power to reject the null hypothesis of the smaller model being correct, when the data are generated from the larger model. Parameter estimation for discrimination designs is also discussed and a simple method is investigated of combining designs to form a hybrid design in order to achieve both model discrimination and estimation. This method has a computational advantage over the use of a compound criterion and the similar performance of the designs obtained from the two approaches is illustrated in an example.     

Multivariate orthogonal series estimates for random design regression

In this paper a new multivariate regression estimate is introduced. It is based on ideas derived in the context of wavelet estimates and is constructed by hard thresholding of estimates of coefficients of a series expansion of the regression function. Multivariate functions constructed analogously to the classical Haar wavelets are used for the series expansion. These functions are orthogonal in _L2(_μn)$L_2({\mu }_n)$, where _μn${\mu }_n$ denotes the empirical design measure. The construction can be considered as designing adapted Haar wavelets.