摘要:

Kadane et ah (1980) describe a method of eliciting a subjective conjugate prior distribution for a linear regression model with the usual normal error structure. Here we describe an experiment in which assessors used a modified form of the method to quantify their opinions about four linear models. The accuracy of assessed prior distributions is measured by a scoring rule and different ways of implementing the method are considered and the effect on scores examined. The empirical results provide guidance for using the method in practice. In addition, the information contained in assessed prior distributions is quantified in terms of equivalent sample sizes, using methods developed here. The question of whether assessed distributions are useful rather than misleading is also addressed

ISSN:0361-0918    

DOI:10.1080/03610919408813206

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

We studied properties of maximum likelihood estimators (MLEs) of the variance components obtained from balanced data of the one-way classification. Exact and asymptotic expected values and variances of these MLEs were derived under the usual normality assumptions. Numerical studies illustrate these expected values and variances, and also illustrate the probability of obtaining a negative solution to the maximum likelihood (ML) equation for the between-class variance component. Simulations were used to study the robustness of the ML estimators under non-normal distributions.

ISSN:0361-0918    

DOI:10.1080/03610919408813207

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Our concern in this paper is a group sequential test design for which the sample sizes between interim analyses are not identical. First, we consider a repeated significance test for comparing two treatments in a clinical trial, and study asymptotic properties of the test statistic. Using the arguments developed by Siegmund (1985, Chapters 8 and 9), we then obtain approximations for the overall significance level of the test and for the error level at each interim analysis. Simulation studies are performed to assess the accuracy of the approximations and the robustness of the approximations are examined using numerical examples.

ISSN:0361-0918    

DOI:10.1080/03610919408813208

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

We investigate the interaction between model selection procedure and testing nonnormality in autoregressive processes when the sample size is small, we utilize residual skewness, kurtosis and an omnibus statistic as test statistics and AIC, SC and HQ as model selection criteria. The result of a Monte Carlo simulation with some AR(3) models shows that the performance of the statistics from the selected model is as well as those from the true mode 1

ISSN:0361-0918    

DOI:10.1080/03610919408813209

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Inference for cross-sectional models using longitudinal data, can be accomplished with generalized estimating equations (Zeger and Liang, 1992). We show that either a diagonal working covariance matrix should be used or a key assumption should be verified. The assumption is non-trivial when covariates vary over time. The validity of this assumption is explored for some broad classes of correlation structures. Similar considerations are shown to be relevant for the more general problem of correlated response data and marginal regression analysis with individual level covariates.

ISSN:0361-0918    

DOI:10.1080/03610919408813210

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

In many circumstances it is necessary to design an experiment with partial confounding of parameter estimates. In this situation, one would like to have tools to assess and compare candidate designs graphically. In this paper we discuss two graphical representations of design properties, and illustrate their application with two examples.

ISSN:0361-0918    

DOI:10.1080/03610919408813211

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

This paper presents a comparative study of the use of Jackknife and Bootstrap resampling procedures to assess the sampling variability of the similarity estimates on the location model when several aspects are varied: the difference between the means, the covariance matrices, the ratio of the generalized' variances, and the corresponding probabilities. The dynamic mechanism between populations derived by the general similarity measures and the assessment of these measures were investigated. A new method was developed to evaluate the similarity measure for location models by modifying the jackknife procedure. Through the simulations, the new method demonstrated a comparable value of bias, variance, and mean squared error of sample estimators of the similarity measure on location model

ISSN:0361-0918    

DOI:10.1080/03610919408813212

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

A multivariate control chart, called M chart, for monitoring if a multivariate process mean has a linear trend over the times at which the subgroups are taken, is proposed. It is shown that dimension-reduction techniques such as projection pursuit would be useful for developing new multivariate control charts.

ISSN:0361-0918    

DOI:10.1080/03610919408813213

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Two stopping rules are defined for the purpose of minimizing the number of iterations needed to provide simulated percentile points with a certain precision: one stopping rule is a result of defining precision relative to the scale of the random variable while the other is a result of defining precision relative to the tail area of the distribution. A simulation experiment is conducted to investigate the effects of the stopping rules as well as the effects of changes in scale. The effects of interest are the precision of the simulated percentile point and the number of iterations needed to achieve that precision. It is shown that the stopping rules are effective in reducing the number of iterations while providing an acceptable precision in the percentile points. Also, increases in scale produce increases in the number of iterations and/or decreases in certain measures of precision

ISSN:0361-0918    

DOI:10.1080/03610919408813214

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

The model proposed by Olkin and Tate (1961) and adopted by Krzanowski (1975) in classifying observations consisting of both binary and continuous data is re-examined when parts of the continuous data are generated from the covariates common to all the underlying populations. The location linear discriminant function adjusted for covariates is constructed. Asymptotic expansion of the distribution of the studentized covariates adjusted discriminant function is derived up to the second order. Performance of the discriminant function can be directly assessed by the plug-in estimator of the overall probability of misclassification derived from the asymptotic expansion.

ISSN:0361-0918    

DOI:10.1080/03610919408813215

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor