The paper presents some approximate and exact tests for testing variance components in general unbalanced mixed linear model. It extends the results presented by Seifert (1992) with emphasis on the computational aspects of the problem.
Hypothesis testing is a model selection problem for which the solution proposed by the two main statistical streams of thought, frequentists and Bayesians, substantially differ. One may think that this fact might be due to the prior chosen in the Bayesian analysis and that a convenient prior selection may reconcile both approaches. However, the Bayesian robustness viewpoint has shown that, in general, this is not so and hence a profound disagreement between both approaches exists. In this paper...
In the paper we prove a formula for the limit of the difference between the power of the asymptotically optimal test and the power of the asymptotically most powerful test for the case of Laplace distribution.
Samples from the gamma population are considered which are censored both above and below, that is, observations below and observations above are missing among observations. The range in such censored samples is taken up and the distribution of this restricted range is obtained, which can be compared with that in the complete sample case given in a previous paper.
Let the random variable have the normal distribution . Explicit formulas for maximum likelihood estimator of are derived under the hypotheses , where are arbitrary fixed numbers. Asymptotic distribution of the likelihood ratio statistic for testing this hypothesis is derived and some of its quantiles are presented.
The problem of testing hypothesis under which the observations are independent, identically distributed against a class of alternatives of regression in a parameter of the one-parameter exponential family is studied. A parametric test for this problem is suggested. The relative efficiency of the parametric test compared to the rank test proposed in the author's preceding paper is also derived.
Approximations to the critical values for tests for multiple changes in location models are obtained through permutation tests principle. Theoretical results say that the approximations based on the limit distribution and the permutation distribution of the test statistics behave in the same way in the limit. However, the results of simulation study show that the permutation tests behave considerably better than the corresponding tests based on the asymptotic critical value.
Consideramos la conexión que existe entre la información de Kullback y los tests admisibles óptimos en el conjunto de riesgos de Neyman-Pearson, usando para ello el estudio de problemas de programación matemática de tipo infinito. Se obtienen resultados que caracterizan un subconjunto de soluciones Bayes como consecuencia del conocimiento de la información, así como una medida de discriminación entre hipótesis para el conjunto de riesgos.
La connaissance de la robustesse des méthodes usuelles d'inférence statistique vis-à-vis de divers types d'écarts par rapport au modèle de base est essentielle pour leur bonne utilisation. Dans cet article sont exposés un certain nombre de résultats (pour la plupart classiques. mais parfois mal connus) concernant la robustesse de ces méthodes vis-à-vis de la non-normalité (pour les comparaisons de moyennes, puis pour les comparaisons de variances), vis-à-vis de la non-équidistribution et de la non-indépendance,...
A method is introduced to select the significant or non null mean terms among a collection of independent random variables. As an application we consider the problem of recovering the significant coefficients in non ordered model selection. The method is based on a convenient random centering of the partial sums of the ordered observations. Based on L-statistics methods we show consistency of the proposed estimator. An extension to unknown parametric distributions is considered. Simulated examples...