摘要:

This article reports progress on the development of a population projection process that emphasizes model selection over demographic accounting. Transparent multiregional/multistate population projections that rely on parameterized model schedules are illustrated, together with simple techniques that extrapolate the recent trends exhibited by the parameters of such schedules. The parameterized schedules condense the amount of demographic information, expressing it in a language and variables that are more readily understood by the users of the projections. In addition, they permit a concise specification of the expected temporal patterns of variation among these variables, and they allow a disaggregated focus on demographic change that otherwise would not be feasible.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478237

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Estimates of the civilian noninstitutional population define the sample frame of most major national sample surveys and are used as denominators in the computation of prevalence rates for various sociodemographic phenomena, such as labor force participation and school enrollment. Current U.S. Bureau of the Census methods for making intercensal estimates and postcensal projections of the civilian noninstitutional population rest on the assumption that the age-, sex-, and race-specific proportions of the population that are institutionalized—as estimated by the last census—remain constant until the next census. This article examines the empirical validity of this assumption by using data from the decennial censuses for 1940–1980 and, in light of substantial decade to decade changes in the age patterns of the institutional proportions for sex- and race-specific populations, seeks to develop alternative methods. To pursue the latter objective, parametric curves are fit to the age-specific institutional proportions for each population for each decade. A study of the observed historical variation in the parameters of these curves then leads to some suggestions about how their shapes can be estimated between censuses and projected beyond the latest available census to provide more accurate estimates and projections of the civilian noninstitutional population.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478238

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

This article examines a method of adjusting subgroup means when there are errors both in the allocation of elements to the subgroups and in the measurement of the response. The advantages with the method are that it is simple, it is also applicable when there are many subgroups, and it requires no distributional assumptions for the response variable. The method is compared to the maximum likelihood method for a dichotomous response, and an example from the Swedish Level of Living Survey is given.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478239

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

This article describes approximations to the posterior means and variances of positive functions of a real or vector-valued parameter, and to the marginal posterior densities of arbitrary (i.e., not necessarily positive) parameters. These approximations can also be used to compute approximate predictive densities. To apply the proposed method, one only needs to be able to maximize slightly modified likelihood functions and to evaluate the observed information at the maxima. Nevertheless, the resulting approximations are generally as accurate and in some cases more accurate than approximations based on third-order expansions of the likelihood and requiring the evaluation of third derivatives. The approximate marginal posterior densities behave very much like saddle-point approximations for sampling distributions. The principal regularity condition required is that the likelihood times prior be unimodal.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478240

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

In this article we introduce a formal procedure for the use of expert opinions in reliability (and fault tree) analysis. We consider the case of multicomponent parallel redundant systems for which there could be a single expert or a group of experts giving us opinions about each component. Inherent in our approach are a procedure for reflecting our judgment of the experts' expertise and our own opinions about the components' life lengths. An important issue here is the induced dependencies between the components' life lengths due to any common knowledge shared by the experts. Our final results are approximations that depend on our having small uncertainty about what the experts have to say. The approximations are easily computable, and they can be generalized to cover any coherent system.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478241

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

We introduce anew better than used of age t0(NBU-t0) class of life distributions, where the survival probability at age 0 is greater than or equal to the conditional survival probability at specified aget0> 0. The dual class ofnew worse than used of age t0(NWU-t0) life distributions is defined by reversing the direction of inequality. Preservation and nonpreservation properties of the two classes under various reliability operations are presented. We also show how to test, using a random sample, whether the NBU-t0class obtains.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478242

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

We consider the penalized likelihood method for estimating nonparametric regression functions in generalized linear models (Nelder and Wedderburn 1972) and present a generalized cross-validation procedure for empirically assessing an appropriate amount of smoothing in these estimates. Asymptotic arguments and numerical simulations are used to show that the generalized cross-validatory procedure preforms well from the point of view of a weighted mean squared error criterion. The methodology adds to the battery of graphical tools for model building and checking within the generalized linear model framework. Included are two examples motivated by medical and horticultural applications.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478243

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

I consider the distribution of Pearson's statistic and of the likelihood-ratio goodness-of-fit statistic for discrete data in the important case where the data are extensive but sparse. It is argued that the appropriate reference distribution is conditional on the sufficient statistic for the unknown regression parameters, β. The first three conditional asymptotic cumulants are derived by Edgeworth expansion, and these are used for the computation of tail probabilities. The principal advantage of the limit considered here, as opposed to the more usualX2limit, is that the cell counts need not be large.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478244

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Given a prediction rule based on a set of patients, what is the probability of incorrectly predicting the outcome of a new patient? Call this probability the true error. An optimistic estimate is the apparent error, or the proportion of incorrect predictions on the original set of patients, and it is the goal of this article to study estimates of the excess error, or the difference between the true and apparent errors. I consider three estimates of the excess error: cross-validation, the jackknife, and the bootstrap. Using simulations and real data, the three estimates for a specific prediction rule are compared. When the prediction rule is allowed to be complicated, overfitting becomes a real danger, and excess error estimation becomes important. The prediction rule chosen here is moderately complicated, involving a variable-selection procedure based on forward logistic regression.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478245

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

An approximate method is given to estimate the mean of a dependent variable after a linear model is fitted to the Box—Cox power transformation of this variable. The estimate is accurate except when the transformation power parameter is near zero. The properties of the estimate in the one-sample and regression cases are considered, by both asymptotic calculations and Monte Carlo simulations, and comparisons are made with the smearing estimate (Duan 1983). It is shown that there can be some cost due to estimating the power transformation, as opposed to assuming it is known; however, this cost is not severe.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478246

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor