Showing papers in "Journal of Multivariate Analysis in 1991"

TL;DR: Breidt and Rosenblatt as mentioned in this paper proposed a maximum likelihood estimation for non-causal autoregressive processes, based on the maximum likelihood estimator for noncausal processes.

TL;DR: In this paper, an estimation problem with observations from a Gaussian process was considered, and it was shown that the maximum likelihood estimator of the identifiable parameter θσ2 is strongly consistent and converges weakly (when normalized by √ n) to a normal random variable, whose variance does not depend on the selection of sample points.

TL;DR: In this article, a class of new parametric models on the unit simplex in R m is introduced, the distributions in question being obtained as conditional distributions of m independent generalized inverse Gaussian random variables given their sum.

TL;DR: In this paper, it was proved that there is a nonparametric estimate T n of the conditional αth quantile of a random vector such that it achieves the optimal rate of convergence n − r in L q -norms (1 ≤ q t n of T n, where t n is a derivative of θ of order m.

TL;DR: In this paper, a Bahadur representation for regression quantiles is provided for error processes which are highly non-stationary (for which there is a nonvanishing bias term) and which are close to being m-dependent.

TL;DR: In this article, it was shown that strongly harmonizable almost periodically correlated processes are almost PC if and only if their spectral measure is concentrated on a countable set of diagonal lines Sλ = {(γ 1, γ 2): γ2 = γ1 − λ; under this same condition, the set Λ = {λ: a(λ, τ) ≠ 0 for some τ} is countable.

TL;DR: In this article, the rate of convergence of various sweep strategies of stochastic relaxation for simulating multivariate Gaussian measures is calculated and compared, where each sweep strategy prescribes a method for chosing which coordinates of the random vector are to be updated.

TL;DR: In this paper, a new form of multivariate gamma is defined whose components are positively correlated and have a three parameter gamma distribution, and explicit forms of moments, moment generating function, conditional moments, and density representations are derived.

TL;DR: In this paper, the authors obtained rates of convergence of stochastic relaxation (heat bath algorithm) for continuous densities which have the form of bounded perturbations of Gaussian densities.

TL;DR: In this article, the authors show the existence of selectors for a multivalued martingale and supermartingale with closed values in a separable Banach space X. The existence of L1(X)-bounded or uniformly integrable selectors is discussed.

TL;DR: In this article, it is shown that all regression estimates based on residuals alone in linear models have zero breakdown slopes in contamination neighborhoods, even though they can have breakdown points as high as one-half.

TL;DR: In this paper, a Cramer-Rao type lower bound for minimum loss unbiased estimators with values in a manifold is derived, and the corresponding notion of efficiency is investigated.

TL;DR: In this paper, an asymptotic distribution theory for signed-rank serial statistics is developed for a general class of serial statistics, and then the authors derive locally optimal tests (in the maximin sense) for testing an ARMA model against other ARMA models.

TL;DR: In this article, the authors investigate nonparametric curve estimation by kernel methods when the observed data satisfy a strong mixing condition and give precise asymptotic evaluations of these errors.

TL;DR: In this paper, three classes of expansions for the distribution function of the χ k 2 (d, R )-distribution are given, where k denotes the dimension, d the degree of freedom, and R the co-correlation matrix.

TL;DR: In a multiparameter situation, the authors characterizes priors under which the Bayesian and frequentist Bartlett corrections for the likelihood ratio statistic differ by o(1) in a multi-dimensional setting.

TL;DR: In this article, a random censorship model is proposed to permit uncertainty in the cause of death assessments for a subset of the subjects in a survival experiment. But only some of the solutions are consistent; i.e., the MLEs and self-consistent estimators are not consistent in general.

TL;DR: For the mean vector of a p-variate normal distribution (p ≧ 3), the generalized Bayes estimators dominating the James-Stein estimator under quadratic loss are given based on the methods of Brown, Brewster and Zidek for estimating a normal variance as mentioned in this paper.

TL;DR: In this paper, a new class of invariant minimax estimators for the case p > m + 1, which are multivariate extensions of the estimators of Stein and Baranchik, is proposed.

TL;DR: In this paper, the third-order asymptotic expansion of the distribution of T ∈ S under a sequence of local alternatives has been derived, which can be applied to multivariate analysis and time series analysis.

TL;DR: In this paper, the authors derived the asymptotic distributions for measures of multivariate skewness and kurtosis defined by Malkovich and Afifi if the underlying distribution is elliptically symmetric.

TL;DR: In this paper, the authors established a central limit theorem for empirical processes indexed by smooth functions and used it to establish the asymptotic normality of semiparametric estimators in time series contexts.

TL;DR: In this paper, the authors considered the problem of estimating a regression function with nonrandom design points and dependent errors and constructed a spline estimate of the regression function and obtained its rate of convergence.

TL;DR: In this article, simple Bartlett-type modifications for a wide class of test statistics, including the efficient score and the likelihood ratio statistics, are proposed to improve the performance of test scores.

TL;DR: In this article, the authors proposed and discussed smooth variants of the standard conditional quantile estimator mn−1(λ | x0), 0 < λ < 1, where mn(y|x0) is a (kernel) estimator of m(y |x0).

TL;DR: In this paper, the authors propose nonparametric estimates of the regression function and its derivative when it is only assumed a weak error's structure, and study their local and global asymptotic behaviour when they observe dependent trajectories.

TL;DR: The authors showed that the coverage error of confidence intervals and level error of hypothesis tests for population quantiles constructed using the bootstrap estimate of sample quantile variance is of precise order n − 1 2 in both one-and two-sided cases.

TL;DR: In this article, a broad subset of the range of the infinitesimal generator of the diffusion is identified, for functions in this set functional central limit theorems and laws of iterated logarithm are derived.

TL;DR: In this article, rates of convergence for the distribution of normalized sample extremes to the appropriate limit distribution were discussed, and it was shown that the rate of convergence depends on the dependence functions and that of the marginals.