摘要:

This article reports the results of an analysis of mortality–air pollution relationships for specification errors using the Ramsey specification error tests. The data for the analysis included information on the total mortality rates and the suspended particulates for 50 SMSA's for 1968 and 1969. The findings indicate that specification errors with such models can be a serious problem and will depend on the model estimated.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479868

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Tables for constructing the uniformly most accurate unbiased confidence interval of level 1 − α for σ12/σ22, where σ12and σ22are the variances of two normal distributions, and for tests of level α of hypotheses specifying the value of σ12/σ22are given for α = .1, .05, .01, .001,n1,n2= 1(1)20(2)30, 36, 45, 60, 90, 180, ∞, whereniis the degree of freedom of the estimate ofσi2. The present tables, more accurate than previous ones, may also be applied to make similar inferences about the ratio of the means of two gamma distributions. Illustrations include applications in archaeology and telephony.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479869

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

This article analyzes the differences between the certainty equivalence (CE) and the sequential updating (SU) approach to the problem of multiperiod decision-making under uncertainty. It is assumed that a single target variable, y, depends linearly on a single control variable,x, and an additive and normally distributed disturbance term,u.It is proved that the expected loss associated with CE is infinite, while that associated with SU is bounded above. Moreover, it is shown that truncation of the distribution of the target variableymay result in the convergence of the expected loss associated with CE.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479870

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

This article is concerned with a generalization of the problem of estimation of median effective dose in bioassay with a normal quantal response curve. A new kind of Bayesian motivated procedure is introduced which leads to a strongly consistent estimator. The convergence is robust in that it holds for abundleof sequences of design vectors—an important feature in a mental testing context where a specified design vector cannot be produced on demand.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479871

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

This article considers the variation of the Secretary Problem where at any stage each of the lastmapplicants are available for employment. We show that the form of the stopping rule which maximizes the probability of employing the best applicant is: stop at the first stage, fromr* onwards, when the relatively best applicant is about to become unavailable. A simple algorithm for evaluatingr* and the probability of employing the best applicant is outlined.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479872

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

We use the Dirichlet distribution in a new application as a tractable model for the variability between personal prior probability vectors over a large population of persons. Given a common realized likelihood function, the resulting population distribution of coherent personal posterior probability vectors is analyzed and found to have mixed moments expressible as a one-dimensional integral. Implications for the study of the evolution of scientific knowledge are discussed.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479873

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

The construction of a suitable rule of allocation in the two-population discrimination problem is considered in the case where there are initially available from the populations II1, II2,n1,n2observations andMunclassified observations. An iterative reclassification procedure based on then1+n3+Mobservations is proposed and found asymptotically optimal whenM→ ∞ andn1andn2are moderately large. The case of finiteMis evaluated by a Monte Carlo experiment which suggests that the proposed procedure, after only one iteration, gives a rule with smaller average risk than the usual rule based on just then1+n2classified observations.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479874

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Some aspects of Bayesian methods of inference relative to switching regression models are analyzed. It is shown that Bayesian confidence intervals and tests may be obtained by using Student'stand χ2distributions. Mean biases and MSE of some Bayesian estimates are compared by Monte Carlo methods with those of the ML estimate. The MSE of the Bayesian estimates were uniformly smaller than those of the ML estimate. An experiment designed by Quandt [11] is also analyzed.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479875

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

In the general linear model with observations not necessarily uncorrelated or homoscedastic, Gauss-Markov regression coefficients are superior to ordinary unweighted least squares in the well known BLU sense if the model is correct. However, it is shown that there is a weaker, but always applicable, minimum overall mean squared error sense in which Gauss-Markov residuals and biased residuals are inferior to ordinary least squares residuals as estimators of possible lack-of-fit in the model. This optimality of ordinary least squares is further illustrated by three other types of results about residuals.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479876

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10479877

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor