|
1. |
Separate Estimation of Primary and Secondary Cancer Preventive Impact: Analysis of a Case-Control Study of Skin Self-Examination and Melanoma |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1381-1387
ColinB. Begg,
Ying Huang,
Marianne Berwick,
Preview
|
PDF (804KB)
|
|
摘要:
Case-control studies increasingly have been used to evaluate the impact of cancer screening strategies. In this context the intent of the screening test has been to reduce cancer mortality by early detection of cancers, permitting curative therapy in some patients who would die of the disease if diagnosis were delayed until the disease was detected clinically. This phenomenon is known as secondary prevention. In an analysis of a case-control study of skin self-examination (SSE) in reducing mortality from melanoma, it was recognized that the exposure (SSE) may encourage the removal of precancerous nevi (moles), thereby reducing cancer incidence (primary prevention). This article describes an analytic strategy for obtaining separate estimates of the primary and secondary preventive impact of the screening practice. The method is focused primarily on resolving the problem of lead-time bias, caused by the artifactual advancement of the time of diagnosis in cases detected by screening.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476706
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
2. |
Parametric Event Sequence Analysis: An Application to an Analysis of Gender and Racial/Ethnic Differences in Patterns of Drug-Use Progression |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1388-1399
Kazuo Yamaguchi,
DeniseB. Kandel,
Preview
|
PDF (1280KB)
|
|
摘要:
This article introduces novel statistical models for the sequence analysis of events. The models are formulated to analyze occurrence, association, and sequencing among events as an extension of log-linear models. A set of parameters characterizes marginal odds and odds ratios of frequencies summed across sequence patterns for each combination of the occurrence/nonoccurrence of events. These parameters are used for the analysis of the occurrence and association of events. Another set of parameters characterizes conditional odds and odds ratios among sequence patterns within each combination of the occurrence/nonoccurrence of events. These parameters are used for the analysis of sequencing of events. The models permit a decomposition of the likelihood function into a marginal likelihood component that includes only parameters for occurrence and association among events and a conditional likelihood component that includes only parameters for sequencing among events. The models are then extended further for regressions with covariates. An application analyzes gender and racial/ethnic differences in patterns of drug use progression. Sequential patterns of initiations and association among initiations are analyzed for three groups of drugs: alcoholic beverages, cigarettes, and marijuana. Findings that cross-validate previous findings based on different datasets and findings that are novel are reported.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476707
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
3. |
Physiological Pharmacokinetic Analysis Using Population Modeling and Informative Prior Distributions |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1400-1412
Andrew Gelman,
Frederic Bois,
Jiming Jiang,
Preview
|
PDF (2312KB)
|
|
摘要:
We describe a general approach using Bayesian analysis for the estimation of parameters in physiological pharmacokinetic models. The chief statistical difficulty in estimation with these models is that any physiological model that is even approximately realistic will have a large number of parameters, often comparable to the number of observations in a typical pharmacokinetic experiment (e.g., 28 measurements and 15 parameters for each subject). In addition, the parameters are generally poorly identified, akin to the well-known ill-conditioned problem of estimating a mixture of declining exponentials. Our modeling includes (a) hierarchical population modeling, which allows partial pooling of information among different experimental subjects; (b) a pharmacokinetic model including compartments for well-perfused tissues, poorly perfused tissues, fat, and the liver; and (c) informative prior distributions for population parameters, which is possible because the parameters represent real physiological variables. We discuss how to estimate the models using Bayesian posterior simulation, a method that automatically includes the uncertainty inherent in estimating such a large number of parameters. We also discuss how to check model fit and sensitivity to the prior distribution using posterior predictive simulation. We illustrate the application to the toxicokinetics of tetrachloroethylene (perchloroethylene [PERC]), the problem that motivated this work.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476708
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
4. |
A Random-Effects Model for Cycle Viability in Fertility Studies |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1413-1422
Haibo Zhou,
ClariceR. Weinberg,
AllenJ. Wilcox,
DonnaD. Baird,
Preview
|
PDF (1639KB)
|
|
摘要:
Models for fertility that take into account the timing of intercourse relative to ovulation are needed to estimate the influence of both endogenous and exogenous factors on human fertility. The classical model assumes that some menstrual cycles are “viable” and some are not, where “viability” is determined by whether hormonal, uterine, and gamete-related factors are favorable to gestation. Within each viable cycle, the various days with intercourse are assumed to act independently; within each nonviable cycle, the days with intercourse can have no effect. Cycle viability for individual cycles is latent in that it is not ascertainable when conception does not occur. The classical model neglects the statistical dependency of outcomes among menstrual cycles within individual couples. Current marginal approaches cannot determine the degree to which heterogeneity in fecundability is biologically based versus the degree to which it is secondary to variation in intercourse behavior from couple to couple. We describe a random-effects model based on assuming that the cycle viability probability varies from couple to couple according to a beta distribution, and we use an EM algorithm to fit the model. The proposed estimating procedure is fully expandable to allow covariate effects on the beta variate. Our method can be applied more generally whenever dependency among Bernoulli trials is induced by a susceptibility state and the outcomes can be observed only in the aggregate. Based on data from a cohort of couples with no known fertility problems who were attempting pregnancy, cycle viability is found to be heterogeneous among couples. Stratification on the presence or absence of prenatal exposure of the woman to her mother's cigarette smoking revealed a statistically significant difference in the two cycle viability distributions. We discuss differences in the interpretation of the beta model compared to the marginal approach based on generalized estimating equations.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476709
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
5. |
Estimation of Median Income of Four-Person Families: A Bayesian Time Series Approach |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1423-1431
Malay Ghosh,
Narinder Nangia,
DalHo Kim,
Preview
|
PDF (1863KB)
|
|
摘要:
This article develops a general methodology for small domain estimation based on data from repeated surveys. The results are directly applied to the estimation of median income of four-person families for the 50 states and the District of Columbia. These estimates are needed by the U.S. Department of Health and Human Services (HHS) to formulate its energy assistance program for low income families. The U.S. Bureau of the Census, by an informal agreement, has provided such estimates to HHS through a linear regression methodology since the latter part of the 1970s. The current method is an empirical Bayes method (EB) that uses the Current Population Survey (CPS) estimates as well as the most recent decennial census estimates updated by the per capita income estimates of the Bureau of Economic Analysis. However, with the existing methodology, standard errors associated with these estimates are not easy to obtain. The EB estimates, when used naively, can lead to underestimation of standard errors. Moreover, because the sample estimates are collected through the CPS every year, there is a very natural time series aspect of the data that is currently ignored. We have performed a full Bayesian analysis using a hierarchical Bayes (HB) time series model. In addition to providing the median income estimates as the posterior means, we have provided also the posterior standard deviations. Included in our model is the information on the median incomes of three- and five-person families as well. In this way a multivariate HB procedure is used. The Bayesian analysis requires evaluation of high-dimensional integrals. We have overcome this problem by using the Gibbs sampling technique, which has turned out to be a very convenient tool for Monte Carlo integration. Also, we have validated our results by comparing them against the 1989 four-person median income figures obtained from the 1990 census. We used four different criteria for such comparisons. It turns out that the estimates obtained by using a bivariate time-series model are the best overall. We use a criterion based on deviances for model selection and also provide a sensitivity analysis of the proposed hierarchical model.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476710
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
6. |
Time-Dependent Hazard Ratio: Modeling and Hypothesis Testing with Application in Lupus Nephritis |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1432-1439
Michal Abrahamowicz,
Todd Mackenzie,
JohnM. Esdaile,
Preview
|
PDF (1369KB)
|
|
摘要:
We investigate the association between duration of untreated disease and survival in lupus nephritis, a rare rheumatologic disease. In this case, as in many other studies of survival, a priori considerations suggest that the effect of the predictor on hazard may change with increasing follow-up time. To accommodate such situations, we use regression splines to model the hazard ratio as a flexible function of time. We propose model-based tests of the hypotheses of hazards proportionality and of no association. We evaluate the accuracy of estimation and inference in simulations and also present analysis of a larger medical data set.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476711
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
7. |
A Semiparametric Transformation Approach to Estimating Usual Daily Intake Distributions |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1440-1449
S.M. Nusser,
A.L. Carriquiry,
K.W. Dodd,
W.A. Fuller,
Preview
|
PDF (1634KB)
|
|
摘要:
The distribution of usual intakes of dietary components is important to individuals formulating food policy and to persons designing nutrition education programs. The usual intake of a dietary component for a person is the long-run average of daily intakes of that component for that person. Because it is impossible to directly observe usual intake for an individual, it is necessary to develop an estimator of the distribution of usual intakes based on a sample of individuals with a small number of daily observations on a subsample of the individuals. Daily intake data for individuals are nonnegative and often very skewed. Also, there is large day-to-day variation relative to the individual-to-individual variation, and the within-individual variance is correlated with the individual means. We suggest a methodology for estimating usual intake distributions that allows for varying degrees of departure from normality and recognizes the measurement error associated with one-day dietary intakes. The estimation method contains four steps. First, the original data are standardized by adjusting for nuisance effects, such as day-of-week and interview sequence. Second, the daily intake data are transformed to normality using a combination of power and grafted polynomial transformations. Third, using a normal components-of-variance model, the distribution of usual intakes is constructed for the transformed data. Finally, a transformation of the normal usual intake distribution to the original scale is defined. The approach is applied to data from the 1985 Continuing Survey of Food Intakes by Individuals and works well for a set of dietary components that are consumed nearly daily and exhibit varying distributional shapes.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476712
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
8. |
A New Perspective on Priors for Generalized Linear Models |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1450-1460
EdwardJ. Bedrick,
Ronald Christensen,
Wesley Johnson,
Preview
|
PDF (1916KB)
|
|
摘要:
This article deals with specifications of informative prior distributions for generalized linear models. Our emphasis is on specifying distributions for selected points on the regression surface; the prior distribution on regression coefficients is induced from this specification. We believe that it is inherently easier to think about conditional means of observables given the regression variables than it is to think about model-dependent regression coefficients. Previous use of conditional means priors seems to be restricted to logistic regression with one predictor variable and to normal theory regression. We expand on the idea of conditional means priors and extend these to arbitrary generalized linear models. We also consider data augmentation priors where the prior is of the same form as the likelihood. We show that data augmentation priors are special cases of conditional means priors. With current Monte Carlo methodology, such as importance sampling and Gibbs sampling, our priors result in tractable posteriors.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476713
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
9. |
The Effect of Improper Priors on Gibbs Sampling in Hierarchical Linear Mixed Models |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1461-1473
JamesP. Hobert,
George Casella,
Preview
|
PDF (2083KB)
|
|
摘要:
Often, either from a lack of prior information or simply for convenience, variance components are modeled with improper priors in hierarchical linear mixed models. Although the posterior distributions for these models are rarely available in closed form, the usual conjugate structure of the prior specification allows for painless calculation of the Gibbs conditionals. Thus the Gibbs sampler may be used to explore the posterior distribution without ever having established propriety of the posterior. An example is given showing that the output from a Gibbs chain corresponding to an improper posterior may appear perfectly reasonable. Thus one cannot expect the Gibbs output to provide a “red flag,” informing the user that the posterior is improper. The user must demonstrate propriety before a Markov chain Monte Carlo technique is used. A theorem is given that classifies improper priors according to the propriety of the resulting posteriors. Applications concerning Bayesian analysis of animal breeding data and the location of maxima of unwieldy (restricted) likelihood functions are discussed. Gibbs sampling with improper posteriors is then considered in more generality. The concept of functional compatibility of conditional densities is introduced and is used to construct an invariant measure for a class of Markov chains. These results are used to show that Gibbs chains corresponding to improper posteriors are, in theory, quite ill-behaved.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476714
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
10. |
The Notion of “Composite Reliability” and its Hierarchical Bayes Estimation |
|
Journal of the American Statistical Association,
Volume 91,
Issue 436,
1996,
Page 1474-1484
Jingxian Chen,
NozerD. Singpurwalla,
Preview
|
PDF (1803KB)
|
|
摘要:
In this article we introduce the notion of “composite reliability” as a measure of the overall reliability of a collection of heterogeneous but similar items. We then propose a hierarchical Bayes model for estimating the composite reliability of systems whose lifelengths are expressed as binary random variables. Subsequently, we develop an alternative approach for the simultaneous inference about many small but related binomial parameters. We propose a two-stage prior for addressing problems of this type and use the Gibbs sampler algorithm for addressing problems involving small proportions. Our topic, though suggested by an issue pertaining to the safety of nuclear power plants, arises in other commonly occurring situations pertaining to consumerism, public policy government regulation. Our development generalizes for situations involving lifelengths that need not be binary.
ISSN:0162-1459
DOI:10.1080/01621459.1996.10476715
出版商:Taylor & Francis Group
年代:1996
数据来源: Taylor
|
|