摘要:

The ranking of paired contestants (players) after a series of contests is difficult when every player does not play every other player. In the 1975 JASA Mark Thompson presented a maximum likelihood solution based on the assumption that the probability of any one player defeating any other is a function only of the difference in their ranks. Here the linear approximation to that likelihood is shown to lead to a nonparametric measure of the efficacy of the ranking, called the net difference in ranks (NDR) , which is the sum of the differences in ranks of the paired players in the observed contests that agree with the ranking minus the sum of the differences in ranks in the observed contests that disagree with the ranking (upsets) . The subject is part of a large literature that has been consolidated by H.A. David in The Method of Paired Comparisons (1963, 1988). The method was introduced by the psychophysicist Fechner in 1860 and has been widely applied to sensory testing,

ISSN:0361-0918    

DOI:10.1080/03610919008812886

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

For estimating the dispersion parameter of the negative binomial distribution, the conditional maximum likelihood estimator is compared with the maximum likelihood and moment estimators. The biases, mean squared errors and the Kullback-Leibler risks of the estimators are examined by simulation studies for a single population and multiple ones with a common parameter. We observe that the conditional maximum likelihood estimator is superior to the others in frequency for a single population as a whole, and it is more definite for multiple populations.

ISSN:0361-0918    

DOI:10.1080/03610919008812887

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

Consider the problem of estimating the mean of a distribution in which the observed data are believed to be highly susceptible to gross errors (i.e., a false reading of the true value of the variate). In such situations, so-called robust statistical techniques have been found to be useful. In a nutshell, the essence of robust location estimation is the downweighting of the influence of outliers, where the outlying status of an observation is determined either by its relative order in the sample or according to its distance from the estimate of location.

ISSN:0361-0918    

DOI:10.1080/03610919008812888

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

Parameter estimates of a new distribution for the strength of brittle fibers and composite materials are considered. An algorithm for generating random numbers from the distribution is suggested. Two parameter estimation methods, one based on a simple least squares procedure and the other based on the maximum likelihood principle, are studied using Monte Carlo simulation. In most cases, the maximum likelihood estimators were found to have somewhat smaller root mean squared error and bias than the least squares estimators. However, the least squares estimates are generally good and provide useful initial values for the numerical iteration used to find the maximum likelihood estimates.

ISSN:0361-0918    

DOI:10.1080/03610919008812889

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

The traditional Goldfeld and Quandt heteroskedasticity pretesting methodology disregards the fact that even though heteroskedasticity may exist, its degree of severity might be such that the OLS estimator would still outperform the 2SAE. It, therefore, produces results inferior to the OLS estimator when the degree of severity of heteroskedasticity is very mild. This paper through Monte Carlo simulations shows that the probabilistic pretesting procedure suggested by Adjibolosoo (1989) is more powerful than the traditional Goldfeld and Quandt heteroskedasticity pretesting methodology at the usual traditionally selected levels of significance

ISSN:0361-0918    

DOI:10.1080/03610919008812890

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

The uncertainty in the estimated posterior multinomial probabilities which results from applying Bayes rule using conditional probabilities estimated from a limited data base is investigated. The first order Taylor series estimator for the variance of the estimated posterior probabilities was developed. A simulation study was conducted to ascertain the accuracy of these approximate variance estimates. In addition, the accuracy of both posterior probability estimates and the variance estimates is evaluated for data bases of various sizes. The method is appropriate for any of the usual medical diagnostic problems in which Bayesian methods are applied

ISSN:0361-0918    

DOI:10.1080/03610919008812891

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

Following an inequality of G. Pólya[1949], certain probability integrals, namelyof the limiting singular normal distribution of a specified multinomial, are shown to be bounded above by integrals Tk(Uk) of the uncorrelated standard multivariate normal density over k-dimensional hyperspheres *Ukof radius Uk=Uk(y) and center (0,…,0). The hypersphere Ukis so chosen that its k-dimensional volume is equal to the k-dimensional volume of a simplex Rk, the image under a diagonalizing transformation, of the simplicial region of integration fkin Φk(y,R). For a symmetric multinomial, Rkturns out to be a regular simplex and explicit formulas for the upper bound in the equicorrelated-equicoordinate case, with common correlation ρ= -1/k,- are derived in terms of (a) normal density and distribution functions and (b) incomplete gamma-function ratio.

ISSN:0361-0918    

DOI:10.1080/03610919008812892

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

The power of the classical .F-test for testing the regression coefficient of a general linear model with elliptic t error variable depends on the degrees of freedom of the t- distribution. In this note it is shown that the power of the F-test based on t-distribution is greater than the normal based test at smaller level of significance.

ISSN:0361-0918    

DOI:10.1080/03610919008812893

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

The classical change point problem is considered, from the invariance point of view. Locally optimal invariant tests are derived for the change in level, when the initial level and the common variance are assumed to be unknown. The tests derived by Chernoff and Zacks (1964) and Gardner (1969), for the change in level, when variance is known, are shown to be locally optimal invariant tests.

ISSN:0361-0918    

DOI:10.1080/03610919008812894

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor

摘要:

A test for homogeneity of several populations against the simple tree alternative is proposed when the observations in various groups are subject to the same pattern of random right-censorship. The test is a generalization of the one proposed by Slivka (1970) which is useful when testing time is expensive so that an early termination of an experiment is desirable. The power of the test is examined in a simulation study.

ISSN:0361-0918    

DOI:10.1080/03610919008812895

出版商:Marcel Dekker, Inc.    

年代:1990

数据来源: Taylor