Showing papers in "Journal of Statistical Planning and Inference in 1993"

TL;DR: In this paper, a modified version of Hooke and Jeeves' pattern search with exact or Monte Carlo moment calculations is used to find I-, D- and A-optimal (or nearly optimal) designs for a wide range of response surface problems.

TL;DR: In this article, the authors developed q-analogues of the Delannoy and Schroder numbers derived from several combinatorial statistics: the number of diagonal steps, the area under the path, and the major index.

TL;DR: A cross-validation procedure is introduced in order to select the smoothing parameter of the kernel distribution function estimate, which leads to a selected parameter which is asymptotically optimal with respect to some quadratic measures of error.

TL;DR: In this paper, the quantile estimator XP from the ANNALS of Statistics 17, 1989, for n→∞, p = pn → 0 and npn→0 (n→ ∞).

TL;DR: In this article, conditions under which a one-point design is optimal in a Bayesian sense under a prior distribution on the parameter were derived for symmetric prior distributions with two support points.

TL;DR: In this article, a class of general quadratic forms in the dependent variable for estimating the variance function in a nonparametric heteroscedastic fixed design regression model is proposed.

TL;DR: This work makes an attempt to modeling the considered sequence of events in time (casual chain) in terms of the general framework of marked point processes, and considering then the associated prediction probabilities.

TL;DR: In this article, an asymptotic representation of the integrated square error for the kernel based estimator of hazard rate in the case of right censored samples is obtained, and it is shown that the bandwidth selected by the data based method of least square cross-validation is as-ymptotically optimal.

TL;DR: In this paper, conditions under which an exact D-optimal design for uncorrelated observations with common variance is also D -optimal for correlated observations were established for regression models with multiple response.

TL;DR: In this paper, the authors introduce an estimate of the Weibull tail coefficient α based on a weighted average of the kn upper order statistics of the sample where n-1kn→0 as n→∞.

TL;DR: It is shown how a simple linearization technique is useful for simplifying computations in a wide variety of problems, including Bayesian robustness problems, by converting a single, nonlinear optimization into a set of linear optimizations.

TL;DR: In this paper, the authors investigated the admissibility of X rss and F rss with respect to squared error loss in the general nonparametric estimation problem, and showed that F Rss is inadmissible relative to integrated squared error losses.

TL;DR: In this article, two new procedures for biased coin design, derived within the Bayesian approach, are presented, one using a formal Bayesian motivation; the other relying on ad hoc justifications.

TL;DR: In this article, the authors show that 0-1 tableaux are also an efficient tool in finding and proving identities involving q and p,q -Stirling numbers of both kinds.

TL;DR: The large sample properties of three tests for no effect in nonparametric regression are investigated in this paper, where the tests can all be represented as weighted sums of squared sample Fourier coefficients.

TL;DR: In this paper, conditions and constructions for optimal nested row and column designs are given for balanced incomplete block designs with nested rows and columns and compared to balanced incomplete blocks with nested columns and rows.

TL;DR: In this paper, the authors consider a situation where one is concerned with the effect of small deviations from a base model, which they express using an e-contamination class, and develop a measure, based on a derivative w.r.t.

TL;DR: A partition-theoretic interpretation of the difference of two successive Gaussian polynomials is provided, namely q − j ([ n + j j ] − [ n +J j −1 ]).

TL;DR: In this article, the authors derive formulas for counting paths in Young's lattice that go up or down by horizontal or vertical strips, which are related to Richard Stanley's theory of differential posets in the special case of Young lattice.

TL;DR: In this article, a mixture of the D1-optimality criterion (minimizing the variance of the estimate for the highest coefficient) and the D-optimal criterion (maximising the volume of the ellipsoid of concentration for the unknown parameter vector) in the polynomial regression model of degree n ⊆ N was considered, and a weighted product of both optimality criteria and explicit solutions were given for the proposed criterion.

TL;DR: In this paper, the central limit theorem for perturbed sample quantiles based on a kernel k and a sequence of window-width an > 0 was studied and necessary and sufficient conditions for these quantiles to hold for the sequence {an}.

TL;DR: In this paper, a random sample of size ni was drawn from an exponential distribution with mean λi, i=1,2, satisfying λ1⩽λ2.

TL;DR: Martin and Eccleston as mentioned in this paper generalize those results to the case that the plots within a block are spatially arranged, and introduce new designs that may be appropriate when the dependence structure has additional symmetries.

TL;DR: In this paper, the authors derived the transient solution of M/M/1 queues under (M, N)-policy by an alternative combinatorial method in which lattice paths with diagonal steps are counted.

TL;DR: In this article, the authors have discussed how combinatoric analysis of lattice paths representing the queueing process leads to the transient solutions without lending into complicated analysis of the techniques used earlier.

TL;DR: The D-optimum designs for canonical polynomials on the q-simplex have been known for linear, quadratic, special cubic and cubic without 3-way effect as mentioned in this paper.

TL;DR: In this article, a lattice path method was used to derive transient solutions for Markovian queueing models, where the model is first discretized and then represented by a random walk path.

TL;DR: In this article, a generalization of the connection between Jeffreys prior and the Kullback-Leibler divergence is proposed as a procedure for generating a wide class of invariant priors of which Jeffreys pre-condition is only one.

TL;DR: In this article, the problem of selecting the population with the largest θ-value and simultaneously estimating the θ -parameter of the selected population is considered for k ≥ 2 independent binomial populations, from which X i ∼ B (n i, θ i ), i = 1,…, k, have been observed.

TL;DR: In this paper, various classes of minimum mean square error (MMSE) estimators are derived in the general linear model and a class of linear biased MMSE estimators, a generalization of optimal ridge estimators is introduced.