Showing papers in "Journal of Statistical Planning and Inference in 1989"

TL;DR: In this article, the authors derive a general theory for concave design critria for non-linear models and then apply the theory to logistic regression and propose designs which formally account for the prior uncertainty in the parameter values.

TL;DR: In this article, the authors extend the concept and technique of quasi-likelihood estimation (Wedderburn, 1974), incorporating possible knowledge of the skewness, kurtosis and higher moments of the underlying distribution.

TL;DR: In order to preserve individual anonymity when estimating the prevalence of a disease or trait possesing a possible social stigma, the authors group the sample to be tested in batches of size k, each batch is first given a screening test.

TL;DR: In this paper, all modes of convergence in L 1 (in probability, almost surely, complete) for the standard kernel regression estimate are equivalent, and they show that all modes are equivalent.

TL;DR: The convergence of the algorithm and its various modifications are the main topic of this paper, based on the simple heuristic idea: to delete, at every step of the iterative procedure, ‘bad’ sets of the supporting points and to include ‘good’ ones in the design.

TL;DR: In this paper, statistical tests for testing composite goodness-of-fit hypotheses on the exponential and bivariate Marshall-Olkin exponential distribution are proposed and investigated, based on the empirical moment generating function and are constructed to be consistent against all alternative distributions.

TL;DR: In this article, the existence of stationary optimal policies for general semi-Markov models with possibly unbounded rewards is proved and the corresponding dynamic programming equations are also derived, and a synthesis and extensions of earlier results are presented.

TL;DR: In this paper, a local adaptation of Rudemo (1982) cross-validation technique leads to data-driven locally selected bandwidths, which provides the estimate for which Mean Integrated Squared Error on some interval is smaller than that of the classical kernel estimate based on the global optimal bandwidth.

TL;DR: In this paper, an approximate likelihood ratio test statistic is proposed for testing the equality of several intraclass correlation coefficients when independent samples are drawn from multivariate normal populations, and the asymptotic null and non-null distributions of the proposed statistic are derived for the nonnormal populations.

TL;DR: In this paper, the convexity of the sets L (Cy + c), A, and A 0 of all linear estimators that are admissible for β under M and M0 with respect to the mean square error criterion are derived via appropriately reducing the sets of L (β^) and L(β^0, where β^ and β^0 are the minimum dispersion linear unbiased estimators of β in these two models.

TL;DR: In this paper, the authors used residuals in the two-sample problem to test the homogeneity of scale in the presence of nuisance location parameters and obtained new asymptotic results for U-statistics with estimated parameters.

TL;DR: In this article, the hazard rate r is defined in terms of ǫ and the survival function F, where F(x)=P(X1 > x), by a straightforward generalization of the definition of r in the one-dimensional case.

TL;DR: A non-asymptotic approach to model selection and the estimation of performance and other parameters affected by the model selection, which treats data based model selection as an integrated part of the estimation procedure.

TL;DR: In this paper, Chen et al. studied a family of test statistics intended for testing exponentiality against DMRL (IMRL), which consists of those life distributions which have decreasing (increasing) mean residual life.

TL;DR: Lower bounds are derived for rates of convergence in estimating the size of jump discontinuities in a regression function or its derivative because the latter rates are nonparametric which indicates that parametric convergence rates are not possible in such instances.

TL;DR: In this article, the authors show that confidence sets recentered at Stein-type estimators have larger coverage probability then the usual confidence ellipsoids, however, the minimum coverage probability of these improved sets is identical to that of the usual sets, so that only 1 − α can be actually reported.

TL;DR: Godambe and Liang as mentioned in this paper extended the results by Godambe (1980, 1984) and Liang (1983) concerning ancillarity of statistic T for parameter of interest θ1 in the presence of parameter θ2, and the loss of information in using the conditional distribution for inference concerning θ 1, and examined relationships concerning the appropriate information matrices.

TL;DR: In this paper, it was shown that for any sequence of positive integers k n → ∞ and k n n → 0 as n → ǫ, the sum of the upper kn extreme values Xn-kn+1,n+n+ … +X>n,n, when properly centered and normalized, converges in distribution to a standard normal random variable N(0,1).

TL;DR: In this article, a dynamic programming equation is derived for general semi-Markov models under the long-run average reward criterion, focusing on it as a limiting case of optimization under discounting as the discount factor goes to one.

TL;DR: In this article, the optimal bandwidth ratio for second-order and fourth-order kernels was derived for the generalized jackknife combination of two quadratics with different bandwidths. And the potential advantages of a fourth-organised kernel make it a viable competitor to the usual secondorder kernel estimator.

TL;DR: In this paper, the authors derive combinatorial conditions which guarantee that the Hausdorff dimension of a set of real numbers equals the entropy of the set of their corresponding r-adic expansions.

TL;DR: In this article, a general model is given and optimal designs are obtained for some polynomial models with certain types of prior distributions over the set of all polynomials, assuming that a prior distribution on possible departures from the assumed model is known.

TL;DR: In this article, an empirical Bayes selection rule was proposed to select the best population under the binomial-beta statistical framework, and it was shown that the Bayes risk converges to the corresponding minimum risk with rates of convergence at least of order O(exp(− cn )) for some positive constant c, where n is the number of accumulated past experience (observations) at hand.

TL;DR: In this paper, the authors proposed a more intuitive and meaningful estimator and study its efficiency compared with other related estimators, such as Bansal and Singh (1985), and suggested that certain characteristics under study are poorly correlated with the selection probabilities.

TL;DR: Asymptotic normality of these statistics is proved for Bernoulli graphs in which the edge probability is constant and it is shown that subgraph counts asymptotically are linear functions of the number of edges in the graph.

TL;DR: In this article, the inverse-partitioned-matrix method is adopted to represent the consistency condition, the minimum dispersion linear unbiased estimators of estimable linear functions of β, and the minimum norm quadratic unbiased estimator of σ 2, all corresponding to the model M merged with linear restrictions Rβ = r, in terms referring explicitly to the unrestricted model M and the restrictions.

TL;DR: In this article, the problem of finding the population which is associated with the largest of the unknown "success probabilities" p1,…,pk is studied under the "0-1" loss, some linear loss which occurs in gambling, and a general monotone, permutation invariant loss.

TL;DR: The optimality result is valid for any Schur criterion and thus holds for the commonly used criteria such as A-, D- or E-optimality as discussed by the authors. But it is not applicable to the class of row-column designs given by Anderson and Eccleston.

TL;DR: In this paper, a simple nonparametric estimator for densities on spheres and a speed of strong uniform convergence are presented. But their estimators are similar to the so-called naive estimators on Euclidean spaces.

TL;DR: In this article, a general sequential procedure for the elimination of inferior elements from a set of m populations is proposed, which consists of applying simultaneously all pairwise comparisons, when it is rejected as being inferior in an appropriate individual test, e.g. in a repeated significance test.