ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482146

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

A method of drawing a single conclusion from the results of a number of independent tests of a statistical hypothesis was pointed out some years ago by Fisher. Modifications of Fisher's method to better adapt it to the case where the underlying probability distributions are discrete have been proposed by Wallis, Lancaster, and Pearson. The relationship between these can be clarified by comparing each of them with a modified form of Pearson's method.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482147

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

This paper provides an evaluation of the effects of a particular regression-estimation technique used to augment measurements when concurrent and additional measurements are available for two other variates. The investigation demonstrates that this use of regression does not always produce better estimates of population parameters, and indeed, that it may produce poorer estimates than could be obtained from the original data alone. Tables of a parameterI, the relative information, are provided so that the utility of making regression estimates to decrease the variance of the mean or variance may be easily evaluated. And finally, the use of these results in the specification of an optimal sampling program is briefly indicated.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482148

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

This paper proves a number of inequalities which improve on existing upper limits to the probability distribution of the sum of independent random variables. The inequalities presented require knowledge only of the variance of the sum and the means and bounds of the component random variables. They are applicable when the number of component random variables is small and/or have different distributions. Figures show the improvement on existing inequalities.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482149

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

It is shown that the BAN (best asymptotically normal) estimates of μ the mean and σ the standard deviation of normal populations based on incomplete (singly or doubly truncated) samples of sizenare asymptotically jointly less efficient than the corresponding estimates based on complete samples of the same size. The joint efficiency,eT(xα,xβ), of the estimates based on incomplete samples is seen to be a monotonic function of the points of truncationxαandxβ(xα<xβ). A table ofeT(xα,xβ) for several given values of the points of truncation is presented.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482150

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

An exact formula for the variance of the product ofKrandom variables,x1,x2, ···,xK, is given as a function of the means and the central product-moments of thexi. The usual approximate variance formula for

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482151

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

Problems of estimating the sum of metric values of a finite population of unknown number of elements are discussed for the case in which independent samples are available. Methods of point estimation applicable to the situation, frequent in economics, in which the samples are not simple random samples but rather are probability samples, with the sampling weights unknown, are investigated. Approximations to the expectation and variance of two alternative estimators are derived, in terms of parameter values. The approximate bias and variance of each is shown to depend strongly on the relation between the sampling patterns and the distribution of the variable. Some discussion of point estimation of the population size and of the mean of the variable when the samples are not simple random samples is given. The entire discussion concerns the use of previously (and unsatisfactorily) collected data, rather than the use of data from well-designed samples.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482152

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482153

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

A distribution-free test of the hypothesis that identically distributed paired observations are independent may be based on a test statistic,D, which measures the variation between the ranks in each pair of observations. Examples of applications of this proposed test are discussed. Further, since the statistic is related to the one used in the Kruskal-WallisH-test applied to samples of 2, some results are of interest for what they suggest about theH-test. Properties of the proposed test under the null hypothesis that are considered include exact sampling distributions, some exact probabilities, moment coefficients, the asymptotic distribution, and approximations to critical values. Large-sample theory for describing alternative hypotheses, for approximating the power of the test, and for determining sample size is also suggested.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482154

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

Linear programming as a problem in optimizing a linear functional subject to linear inequality constraints is first discussed along with possible uses of the resulting duality relations. Various approaches to dealing with such problems when parts of the data are subject to error are briefly reviewed. Chance constrained programming refers to the class of such cases in which contraint violations are admissible up to pre- assigned probability levels. This topic is elaborated in the context of a ship-chartering problem in which independent normally distributed deviates from the (known) average demands may occur in any of the periods to be considered. Certain additional assumptions (including the use of a specified class of linear decision rules) make it possible to effect reductions first to a nonlinear (mathematical) programming and then to a linear programming problem. The use of the resulting duality relations for evaluating risk and quality levels of planned performance are then briefly examined.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10482155

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor