摘要:

The concept of conditional experimental frames of reference has a significance for the general theory of statistical inference which has been emphasized by R. A. Fisher, D. R. Cox, J. W. Tukey, and others. This concept is formulated as aprinciple of conditionality, from which some general consequences are deduced mathematically. These include thelikelihood principle, which has not hitherto been very widely accepted, in contrast with the conditionality concept which many statisticians are inclined to accept for purposes of “informative inference.” The likelihood principle states that the “evidential meaning” of experimental results is characterized fully by the likelihood function, without other reference to the structure of an experiment, in contrast with standard methods in which significance and confidence levels are based on the complete experimental model. The principal writers supporting the likelihood principle have been Fisher and G. A. Barnard, in addition to Bayesian writers for whom it represents the “directly empirical” part of their standpoint. The likelihood principle suggests certain systematic reinterpretations and revisions of standard methods, including “intrinsic significance and confidence levels” and “intrinsic standard errors,” which are developed and illustrated. The close relations between non-Bayesian likelihood methods and Bayesian methods are discussed.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480660

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480661

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

The following generalization of the classical ballot theorem is given: Suppose that an urn containsncards marked with nonnegative integers whose sum isk≦n. All thencards are drawn without replacement from the urn. The probability that forr= 1, 2, ···,nthe sum of the firstrnumbers drawn is less than r is 1 —k/n. By using the ballot theorem and its generalization the author findsGn(x) the probability that a busy period consists of servingncustomers and its length is ≦xfor single-server queues when either the inter-arrival times or the service times have an exponential distribution. Finally, the author gives the general solution of the classical ballot problem as well as an application of it in the theory of Bernoulli trials.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480662

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

An examination of tables from the 1950 U. S. Census of Population, and of the basic Persons punch card, shows that a few of the cards were punched one column to the right of the proper position in at least some columns. The result is that numbers reported in certain rare categories—very young widowers and divorcés, and male Indians 10–14 or 20–24—were greatly exaggerated. These errors occurred in spite of a careful checking program, and illustrate the necessity for users to view data concerning rare categories with special caution.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480663

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

In this paper a method of estimating the parameters of a set of regression equations is reported which involves application of Aitken's generalized least-squares [1] to the whole system of equations. Under conditions generally encountered in practice, it is found that the regression coefficient estimators so obtained are at least asymptotically more efficient than those obtained by an equation-by-equation application of least squares. This gain in efficiency can be quite large if “independent” variables in different equations are not highly correlated and if disturbance terms in different equations are highly correlated. Further, tests of the hypothesis that all regression equation coefficient vectors are equal, based on “micro” and “macro” data, are described. If this hypothesis is accepted, there will be no aggregation bias. Finally, the estimation procedure and the “micro-test” for aggregation bias are applied in the analysis of annual investment data, 1935–1954, for two firms.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480664

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

When interdependence of disturbances is present in a regression model, the pattern of sample residuals contains information which is useful in prediction of post-sample drawings. This information, which is often overlooked, is exploited in the best linear unbiased predictor derived here. The gain in efficiency associated with using this predictor instead of the usual expected value estimator may be substantial.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480665

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

For a queue with many servers and constant holding time, the distribution of the total service time for a fixed observation interval during a period of statistical equilibrium has been derived. The derivation, while peculiar to the special case where holding time is assumed constant, involves elementary probabilistic arguments. An application is given in which the distribution of the total service time for a fixed observation interval is used to solve a problem in plumbing design.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480666

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

Sequential selection techniques are presented which can be utilized to determine items for inclusion in samples as if these items were selected by classical, non-sequential, sampling plans. These techniques eliminate the need to sort data according to classification categories prior to sample selection. Thus, these techniques will enable effective use of digital computers to select samples according to most known sampling plans. General computer programs which will utilize these sequential selection techniques are also described.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480667

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

The confounding of effects and of interactions in experiments in the 2p–qseries is usually resolved by doing other fractions of the same size as the first one. This paper gives augmenting fractions which require fewer runs. If only one ambiguity is present, it is possible thatonemore run may resolve it. If more than one ambiguity is suspected, then more runs will be required.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480668

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor

摘要:

The technique of latent class analysis has not previously been applied to demographic data. The present paper analyzes 3709 admissions to a hospital for the retarded on the basis of five variables, and shows how a latent class analysis reduces the population to two groups, each of which tend to feature within group independence among the five variables. A mortality analysis of each group shows great differences in the mortality rate when length of stay in the hospital is held constant. It is concluded that latent class analysis has possibilities as an alternative standardization process for demographic analysis.

ISSN:0162-1459    

DOI:10.1080/01621459.1962.10480669

出版商:Taylor & Francis Group    

年代:1962

数据来源: Taylor