摘要:

Considerk(k≥ 2) increasing failure rate (IFR) populations which differ only in their scale parameters. A multiple decision approach is taken to the problem of ranking and selection of these populations with respect to the means. The procedures are based on the wellknown statistic of Epstein and Sobel [4]. It is shown that the probability of correction selection using these procedures is greater than or equal to the corresponding probability of correct selection for exponential populations. Hence, exact but somewhat conservative solutions are obtained.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481503

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

A single-sample procedure is proposed for obtaining asymptotically optimum confidence intervals for the largest mean ofknormal populations with common unknown variance.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481504

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

Suppose μi, σi2are the mean and variance of theith normal population,i= 1, 2, 3 (we assume the parameters to be all unknown). For given nonzero constants λ1, λ2, λ3, suppose μ = Σ3i=1λiμi. Sequential procedures have been derived (i) to arrive at a fixed width confidence interval for μ, and (ii) to estimate μ pointwise with minimum risk (loss being squared error). Some desirable asymptotic properties are proved for the stopping times in both cases. Monte Carlo studies were made for the problem in (i) with moderate optimal sample sizes.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481505

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

Often in reporting the results ofaregression analysis, researchers, particularly in the social sciences, choose to standardize the estimators of the regression coefficients into what are called “beta coefficients.” Most studies in which beta coefficients are reported involve linear models which contain stochastic predictor variables. However, in dealing with regression models in which thep redictor variables are nonstochastic, standardized regression coefficients can be defined which are analogous to those found in the models with stochastic predictor variables. We consider the problem of estimating these parameters. Several estimators are introduced and their properties are discussed. Two data examples are included to demonstrate the empirical behavior of the various estimators.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481506

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

Longley [10] proposedaset of data for use in testing regression programs. This paper shows that the numerically accurate solution in this example was probably an unreasonable estimate of the regression coefficients. This is true because the accuracy of the data and appropriateness of the model may affect the solution more than the computational method. An easily computed index is derived that can be used to indicate such computational instability. The basic conclusion is that a concern about highly accurate computational methods must be tempered with a concern for whether the data are accurate enough to make the results meaningful.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481507

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

The method of iterative weighted least squares can be used to estimate the parameters in a nonlinear regression model. If the dependent variables are observations from a member of the regular exponential family, then under mild conditions it is shown that the IWLS estimates are identical to those obtained using the maximum likelihood principle. An application is provided to illustrate the results.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481508

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481509

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

A two-stage estimator is proposed for the cell probabilities in a two-dimensional contingency table. Using a multivariate mean square error criterion, we show that the decision to use either the observed cell probabilities or the estimators assuming row-column independence is approximately dependent upon the noncentrality parameter of the Pearson chi-square statisticx2. We then derive a decision rule that says to use the independence-assumption estimators whenX2is less than twice the number of degrees of freedom for testing independence.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481510

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481511

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor

摘要:

In this article we apply the group-testing idea to the estimation of a percentile of a normal population when both the mean and the variance are unknown. The structure is such that we observe the sum of the normal, independent measurements ofmunits in one single observation (or test). Under the assumption of a fixed total costT, we have obtained the optimal value of the group sizemfor both the small-sample and the large-sample (T→ ∞) cases. Comparison with the un-grouped method (m= 1) is made and the improvement is significant.

ISSN:0162-1459    

DOI:10.1080/01621459.1976.10481512

出版商:Taylor & Francis Group    

年代:1976

数据来源: Taylor