论文检索

摘要暂缺

摘要暂缺

摘要暂缺

摘要暂缺

均值变化最大似然检验的渐近分布AI 解读

For the maximum likelihood tests for a change in the mean of independent random variables, it is proved that the limit distribution is a double exponential, i.e. Gumbel, distribution. Our results also explain why the maximum likelihood methods are very powerful on the tails.

摘要暂缺

一类允许长程依赖的随机场模型AI 解读

In this note it is shown how a convenient class of random field models with rational spectral densities can be augmented to incorporate the possibility of long range dependence.

比率策略的渐近极小极大性AI 解读

The use of simple random sampling together with ratio estimation is of fundamental importance in survey sampling practice. From a theoretical point of view the ratio estimator has been justified by Bayesian methods (Ericson, 1969) and by superpopulation methods (Royall, 1970). But these methods do not yield a rigorous justification of simple random sampling at the same time. We use the minimax principle in an asymptotic version to justify both simple random sampling and…

两次调查中部分替换单位的连续抽样贝叶斯方法AI 解读

To study time-dependent characteristics of a population, we may sample the same population at regular intervals. A basic problem is how the sample should be selected at different time periods. Here we consider successive sampling with partial replacement of units on two occasions from a Bayesian point of view. Bayes estimators and optimal designs for such priors will be obtained. In one situation, the whole survey, i.e. the first sample and the second sample, is planned…

核回归估计中的权重选择AI 解读

For kernel regression estimation a weighting scheme due to Nadaraya and Watson has been associated with random design, and a convolution type weighting scheme with fixed design. Based on integrated mean square error, none of the estimators is uniformly optimal in either design. However, the convolution type weights are minimax optimal. Further advantages of this estimator can be seen in the structure of the bias.

摘要暂缺

边际齐性模型设计矩阵的构造AI 解读

Suppose that a homogeneous group of n subjects is observed on T occasions. Each subject has one of K levels of response at each of the T occasions. The cross-classification of the recorded responses over time forms a KT contingency table. This paper discusses the design matrix for the marginal homogeneity model, which assumes that the first-order marginal probabilities at each time are equal. The marginal homogeneity model is a linear model for the vector of cell…

基于似然的显著性检验的条件性质AI 解读

This paper considers the conditional properties of the unconditional significance tests based on standardized versions of the maximum likelihood estimate and the score statistic for the case of a simple null hypothesis and a one-parameter model. The properties of the tests are assessed by considering the conditional level of the test given the observed value of a statistic that is locally ancillary near the null model. It is shown that for the test based on the maximum…

非正态线性回归：高维显著性水平计算的一个例子AI 解读

Analysis of nonnormal linear models leads to an initial conditioning on the standardized residuals, giving an unnormed density on Rk, where k is the number of parameters. To obtain an observed level of significance for a single parameter it is then necessary to calculate a marginal probability, thus requiring integration in k dimensions. In this paper a conditional approach to evaluating the observed level of significance is developed, and an importance sampling technique is…

多响应回归分析中的近似推断AI 解读

We present an approximate inference procedure for the multiresponse regression model. This can be used to determine approximate confidence regions or intervals and to formulate diagnostic tools such as standardized residuals and influence measures. The overall approximation is based on approximations at two stages. At the first stage, the multiresponse regression model is replaced by a generalized least-squares model using the estimated variance-covariance matrix of the…

摘要暂缺

定位伯努利变量序列中的变化区段AI 解读

The theory underlying inferences about a single change point in a sequence of independent Bernoulli variables is extended to the two change points problem where the distance between the two change points is known, namely a changed segment. Both null and nonnull distributions of the log likelihood ratio for testing the hypothesis that there is no changed segment against the alternative that there is one of specified length are derived using recurrence equations. The…

摘要暂缺

条件逻辑回归似然分析中的鞍点修正AI 解读

Barndorfi-Nielsen & Cox (1979) introduced the double saddlepoint approximation to the conditional likelihood function for a structural parameter vector given a sufficient statistic for a nuisance parameter vector. This approximation is considered in some detail for the case of multiple logistic regression models with polytomous outcome states. It is shown that the profile score with saddlepoint correction yields a simple derivation of an approximation to the mean of a…

重复流感暴发中两种依赖来源的分离AI 解读

Two or more responses measured on a randomly selected individual can be dependent for two quite different reasons. In the first place they can be dependent within, that is conditional upon, each individual. Secondly, dependence is created simply from the heterogeneity between individuals. This paper addresses a rare instance of being able to measure separately the dependence from the two sources. The data comprise frequencies of the infection profiles for a sample of…

多元极值分布建模AI 解读

Multivariate extreme value distributions arise as the limiting joint distribution of normalized componentwise maxima/minima. No parametric family exists for the dependence between the margins. This paper extends to more than two variables the models and results for the bivariate case obtained by Tawn (1988). Two new families of physically motivated parametric models for the dependence structure are presented and are illustrated with an application to trivariate extreme sea…

摘要暂缺

摘要暂缺

利用估计得分进行假设检验AI 解读

A method is proposed for testing hypotheses that is very much in the spirit of expected maximum log likelihood estimation and bears much the same relation to it as the likelihood ratio test methodology does to maximum likelihood estimation. For a single test parameter, the method amounts to basing critical regions on the difference between the estimated score and its expectation. In some cases where a sufficient statistic, independent of the test parameter, exists for the…