ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476762

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

The ecological theory of limiting factors holds that the observed level of response in a biological process will be governed by the input factor in least supply—the limiting factor. This theory has formed the basis for numerous attempts by aquatic ecologists to describe the relation between the biological productivity of inland waters and the availability of plant nutrients required for algal growth. Regression analysis has been the primary statistical tool used in the development of such relations, yet any statistical model that represents the limiting effect of some explanatory factor as an expectation contradicts the substantive theory of limiting factors. Limnological data not resulting in an adequate regression of chlorophyll on phosphorus have been viewed as failing to support the limiting effect of this nutrient on algal biomass in lakes. But when represented by a more appropriate model, such data may be seen to provide similar evidence for the relation of chlorophyll to phosphorus as does data resulting in a strong regression. Data from limnological studies often exhibit a scatter of points distributed in the shape of a triangle lying beneath an upper boundary. Appropriate models for such data are introduced to describe the upper boundary or potential limit, the distribution of points falling below the limit, and the degree of random error. An application of the EM algorithm provides marginal maximum likelihood estimates of the parameters in the more complex models considered. Several results are given for the models, including a goodness-of-fit diagnostic and estimation of the large-sample parameter covariance matrix. Application of the models is illustrated by fitting empirical relationships between chlorophyll and the plant nutrient phosphorus in temperate lakes.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476763

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Screening of pooled urine samples was suggested during the Second World War as a method for reducing the cost of detecting syphilis in U.S. soldiers. Recently, pooling has been used in screening for human immunodeficiency virus (HIV) antibody to help curb the further spread of the virus. Pooling reduces the cost but also—and probably more importantly—offers a feasible way to lower the error rates associated with labeling samples when screening low-risk HIV populations. For example, given the limited precision of the presently available test kits, when the screened population has a prevalence of 4 per 1,000 (which is roughly the estimated U.S. prevalence), the probability that a sample labeled positive is antibody-free can be reduced from approximately 90% (if each sample is tested individually) to about 2%. Furthermore, screening pooled sera samples can also be used to reduce the probability that a sample labeled negative in fact has antibodies up to 40-fold in such a population—an important consideration when attempting to preserve the integrity of the blood supply. In this article we generalize some commonly used pooling procedures and introduce some new ones. We discuss and compare these procedures' relative merits and weaknesses under practical situations in the context of HIV screening. The proposed procedures can also be used and generalized to other medical and quality control applications.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476764

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

We consider the problem of critiquing prior beliefs concerning the distribution of a discrete random variable in the light of a sequentially obtained sample. A topical application concerns a probabilistic expert system for the diagnosis of congenital heart disease, which requires specification of a large number of conditional probabilities that are initially imprecisely estimated by a suitable “expert.” These prior beliefs may be formally updated as data become available, but it would seem essential to contrast the original expert assessments with the data obtained to quickly identify inappropriate subjective inputs. We consider both Bayes factor and significance testing techniques for such a prior/data comparison, both in nonsequential and sequential forms. The common basis as alternative standardizations of the logarithm of the predictive ordinate of the observed data is emphasised, and a Bayesian discrepancy statistic with a variety of interpretations provides a formal means of discounting the expert's judgments in the light of the data. The judgments are found to be of generally high quality, and procedures for automatic monitoring and adaptation are recommended.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476765

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

In the early 1980s, the U.S. Army opened the National Training Center (NTC) as its main U.S. training venue for battalions and brigades. The NTC's primary purpose was training, but because the simulated combat was fairly realistic, many hoped it would provide a sort of laboratory for identifying problems and exploring ways to fix them. This article describes a project that studied tactical reconnaissance at the NTC. The project's substantive purpose was to look for problems in the organization, doctrine, equipment, and training for tactical reconnaissance, and to suggest remedies. Its methodological purpose was to learn how to measure reconnaissance activities and how to use those measurements for analytical purposes, neither of which had been done previously. This article gives background on the NTC and on tactical reconnaissance, and then traces the steps of the reconnaissance project. The two main methodological issues are how to gather measurements in venues like the NTC and what kinds of questions such venues can be used to answer. As it turns out, high-tech measurements are generally of little use. Low-tech measurements are useful for diagnosis (description) and hypothesis generation, predictions and causal inferences being fraught with hazard. Modest as these uses are, the NTC data are nonetheless highly valuable.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476766

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Gamma-ray bursts are short but powerful pulses of electromagnetic radiation coming from space. Specially equipped satellites have observed the arrival of several hundred bursts during the past two decades. The origin of these bursts, whether from near the earth, across the galaxy, or out of the deep cosmos, has become a major question for astronomers. At the heart of this question lies a statistical puzzle relating to familiar issues in survival analysis: Kaplan-Meier-type curves, data truncation, hazard rate analysis, Mantel-Haenszel statistics, and so on. Survival analysis is used here to critically examine two different models that have been proposed for the gamma-ray burst data. The analysis contradicts one of these models.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476767

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Missing data refers to a class of problems made difficult by the absence of some portions of a familiar data structure. For example, a regression problem might have some missing values in the predictor vectors. This article concerns nonparametric approaches to assessing the accuracy of an estimator in a missing data situation. Three main topics are discussed: bootstrap methods for missing data, these methods' relationship to the theory of multiple imputation, and computationally efficient ways of executing them. The simplest form of nonparametric bootstrap confidence interval turns out to give convenient and accurate answers. There are interesting practical and theoretical differences between bootstrap methods and the multiple imputation approach, as well as some useful similarities.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476768

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476769

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476770

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Multidimensional Wilson-Hilferty transformations support Gaussian approximations to certain joint distributions of quadratic forms in jointly Gaussian variates. Central and noncentral joint distributions are studied and applications are noted. Parameters of the approximating distributions are given up to terms of ordero(v−2) in the degrees of freedomv.Numerical studies validate using these approximations over a range of parameters for an essential subclass of the distributions studied, especially in upper tails.

ISSN:0162-1459    

DOI:10.1080/01621459.1994.10476771

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor