Showing papers in "Biometrics in 1991"

Applied Logistic Regression.

Abstract: \"A new edition of the definitive guide to logistic regression modeling for health science and other applicationsThis thoroughly expanded Third Edition provides an easily accessible introduction to the logistic regression (LR) model and highlights the power of this model by examining the relationship between a dichotomous outcome and a set of covariables. Applied Logistic Regression, Third Edition emphasizes applications in the health sciences and handpicks topics that best suit the use of modern statistical software. The book provides readers with state-of-the-art techniques for building, interpreting, and assessing the performance of LR models. New and updated features include: A chapter on the analysis of correlated outcome data. A wealth of additional material for topics ranging from Bayesian methods to assessing model fit Rich data sets from real-world studies that demonstrate each method under discussion. Detailed examples and interpretation of the presented results as well as exercises throughout Applied Logistic Regression, Third Edition is a must-have guide for professionals and researchers who need to model nominal or ordinal scaled outcome variables in public health, medicine, and the social sciences as well as a wide range of other fields and disciplines\"--

Designing Experiments and Analysing Data: A Model Comparison Perspective.

Abstract: Contents: Preface. Part I: Conceptual Bases of Experimental Design and Analysis. The Logic of Experimental Design. Introduction to the Fisher Tradition. Part II: Model Comparisons for Between-Subjects Designs. Introduction to Model Comparisons: One-Way Between-Subjects Designs. Individual Comparisons of Means. Testing Several Contrasts: The Multiple-Comparison Problem. Trend Analysis. Two-Way Between-Subjects Factorial Designs. Higher Order Between-Subjects Factorial Designs. Designs With Covariates: ANCOVA and Blocking. Designs With Random or Nested Factors. Part III: Model Comparisons for Designs Involving Within-Subjects Factors. One-Way Within-Subjects Designs: Univariate Approach. Higher-Order Designs With Within-Subjects Factors: Univariate Approach. One-Way Within-Subjects Designs: Multivariate Approach. Higher Order Designs With Within-Subjects Factors: Multivariate Approach. Part IV: Alternative Analysis Strategies. An Introduction to Multilevel Models for Within-Subjects Designs. An Introduction to Multilevel Hierarchical Mixed Models: Nested Designs. Appendices: Statistical Tables. Part 1. Linear Models: The Relation Between ANOVA and Regression. Part 2. A Brief Primer of Principles of Formulating and Comparing Models. Notes. Solutions to Selected Exercises. References.

Probabilités, Analyse des données et statistique

Abstract: Table des matieres : I. Outils probabilistes. 1. Modele probabiliste. 2. Variables aleatoires. 3. Couples de variables aleatoires, conditionnement. 4. Vecteurs aleatoires. Formes quadratiques et lois associees. II. Statistique exploratoire. 5. Description unidimensionnelle de donnees numeriques. 6. Description bidimensionnelle et mesures de liaison entre variables. 7. L'analyse en composantes principales. 8. L'analyse canonique et la comparaison de groupes de variables. 9. L'analyse des correspondances. 10. L'analyse des correspondances multiples. 11. Methodes de classification. III. Statistique inferentielle. 12. Distributions des caracteristiques d'un echantillon. 13. L'estimation. 14. Les tests statistiques. 15. Methodes de Monte-Carlo et de reechantillonnage (Jack-knife, bootstrap). IV. Modeles predictifs. 16. La regression simple. 17. La regression multiple et le modele lineaire general. 18. Analyse discriminante et regression logistique. 19. Methodes algorithmiques, choix de modeles et principes d'apprentissage. V. Recueil des donnees. 20. Sondages. 21. Plans d'experiences. Annexes. Bibliographie. Index des noms. Index.

Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses.

Abstract: Generalized estimating equations are introduced in an ad hoc fashion for the covariance matrix of a multivariate response. These equations are to be solved jointly with score equations from a generalized linear model for mean parameters. A class of quadratic exponential models is used to develop joint estimating equations for mean and covariance parameters in a more systematic fashion, and proposals for the use of such equations are developed. Comments on the relative merits of the ad hoc and model-based approaches to estimation are given and a regression illustration with a bivariate response is provided.

Some practical aspects of a conditional likelihood approach to capture experiments

Abstract: The use of conditional likelihood methods in the analysis of capture data allows the modeling of capture probabilities in terms of observable characteristics of the captured individuals and the trapping occasions. The resulting models may then be used to estimate the size of the population. Here the use of conditional likelihood procedures to construct models for capture probabilities is discussed and illustrated by an example.

Two guidelines for bootstrap hypothesis testing

Abstract: Two guidelines for nonparametric bootstrap hypothesis testing are highlighted. The first recommends that resampling be done in a way that reflects the null hypothesis, even when the true hypothesis is distant from the null. The second guideline argues that bootstrap hypothesis tests should employ methods that are already recognized as having good features in the closely related problem of confidence interval construction. Violation of the first guideline can seriously reduce the power of a test. Sometimes this reduction is spectacular, since it is most serious when the null hypothesis is grossly in error. The second guideline is of some importance when the conclusion of a test is equivocal. It has no direct bearing on power, but improves the level accuracy of a test.

Regression with frailty in survival analysis.

Abstract: In studies of survival, the hazard function for each individual may depend on observed risk variables but usually not all such variables are known or measurable. This unknown factor of the hazard function is usually termed the individual heterogeneity or frailty. When survival is time to the occurrence of a particular type of event and more than one such time may be obtained for each individual, frailty is a common factor among such recurrence times. A model including frailty is fitted to such repeated measures of recurrence times.

Second-order analysis of spatial clustering for inhomogeneous populations.

Abstract: Motivated by recent interest in the possible spatial clustering of rare diseases, the paper develops an approach to the assessment of spatial clustering based on the second-moment properties of a labelled point process. The concept of no spatial clustering is identified with the hypothesis that in a realisation of a stationary spatial point process consisting of events of two qualitatively different types, the type 1 events are a random sample from the superposition of type 1 and type 2 events. A diagnostic plot for estimating the nature and physical scale of clustering effects is proposed. The availability of Monte Carlo tests of significance is noted. An application to published data on the spatial distribution of childhood leukaemia and lymphoma in North Humberside is described.

Spatial analysis of field experiments : an extension to two dimensions

Abstract: The one-dimensional spatial analysis procedure proposed by Gleeson and Cullis (1987, Biometrics 43, 277-288) is extended to two dimensions using the subclass of separable lattice processes to model the errors. Residual maximum likelihood estimation of the models is described and diagnostics for testing model adequacy are derived. Results from the analysis of 24 sets of uniformity data indicate the frequent need for a two-dimensional analysis even when the plot shape is highly rectangular. These results also indicate the potential gain from using a two-dimensional spatial analysis rather than a row + column analysis. An example is presented of the analysis of a field experiment on tobacco.

A Monte Carlo method for Bayesian inference in frailty models.

Abstract: Many analyses in epidemiological and prognostic studies and in studies of event history data require methods that allow for unobserved covariates or "frailties." Clayton and Cuzick (1985, Journal of the Royal Statistical Society, Series A 148, 82-117) proposed a generalization of the proportional hazards model that implemented such random effects, but the proof of the asymptotic properties of the method remains elusive, and practical experience suggests that the likelihoods may be markedly nonquadratic. This paper sets out a Bayesian representation of the model in the spirit of Kalbfleisch (1978, Journal of the Royal Statistical Society, Series B 40, 214-221) and discusses inference using Monte Carlo methods.

Practical implications of modes of statistical inference for causal effects and the critical role of the assignment mechanism.

Abstract: Causal inference in an important topic and one that is now attracting serious attention of statisticians. Although there exist recent discussions concerning the general definition of causal effects and a substantial literature on specific techniques for the analysis of data in randomized and nonrandomized studies, there has been relatively little discussion of modes of statistical inference for causal effects. This presentation briefly describes and contrasts four basic modes of statistical inference for causal effects, emphasizes the common underlying causal framework with a posited assignment mechanism, and describes practical implications in the context of an example involving the effects of switching from a name-brand to a generic drug. A fundamental conclusion is that in such nonrandomized studies, sensitivity of inference to the assignment mechanism is the dominant issue, and it cannot be avoided by changing modes of inference, for instance, by changing from randomization-based to Bayesian methods. INTRODUCTION Causal Inference Causal inference is a topic that statisticians are addressing more vigorously and rigorously in recent years. This is a desirable development for statistics, as supported by Cox's (1986) comment on Holland (1986b) that “ the issues explicitly and implicitly raised by the article seem to me more important for the foundations of our subject than the discussion of the nature of probability ”.

A goodness-of-fit test for binary regression models, based on smoothing methods

Abstract: A new global test statistic for models with continuous covariates and binary response is introduced. The test statistic is based on nonparametric kernel methods. Explicit expressions are given for the mean and variance of the test statistic. Asymptotic properties are considered and approximate corrections due to parameter estimation are presented. Properties of the test statistic are studied by simulation. The goodness-of-fit method is illustrated on data from a Dutch follow-up study on preterm infants. Recommendations for practitioners are given.

Biostatistics: A Foundation for Analysis in the Health Sciences, 5th Edition.

Abstract: The ability to analyze and interpret enormous amounts of data has become a prerequisite for success in allied healthcare and the health sciences. Now in its 11 th edition, Biostatistics: A Foundation for Analysis in the Health Sciences continues to offer in-depth guidance toward biostatistical concepts, techniques, and practical applications in the modern healthcare setting. Comprehensive in scope yet detailed in coverage, this text helps students understand—and appropriately use—probability distributions, sampling distributions, estimation, hypothesis testing, variance analysis, regression, correlation analysis, and other statistical tools fundamental to the science and practice of medicine.

Universal kriging and cokriging as a regression procedure.

Abstract: Prediction of a property on the basis of a set of point measurements in a region is required if a map of this property for the region is to be made. Of the spatial interpolation and prediction techniques, kriging is optimal among all linear procedures, as it is unbiased and has minimal variance of the prediction error. In cokriging, which has this same attractive property, additional observations of one or more covariables are used, which may lead to increased precision of the predictions. Both techniques are often applicable in different fields such as soil science, meteorology, medicine, agriculture, biology, public health, and environmental sciences (e.g., atmospheric or soil pollution). In this study we try to remove the cloud of obscurity covering the notions of kriging and cokriging by embedding them into regression procedures. This leads to a straightforward formulation of the two techniques. It turns out that kriging and cokriging differ only slightly from each other. The procedures are illustrated by two numerical examples, one to demonstrate the methodology, and one practical problem encountered in a soil study. Cokriging is found to be most valuable when a highly correlated covariable is sampled intensely.

A random field approach to the analysis of field-plot experiments and other spatial experiments

Abstract: Several "nearest-neighbor" methods for the analysis of data from spatial experiments (e.g., agricultural field experiments) have recently been proposed. These methods attempt to account for the effect of spatial heterogeneity on the estimation of treatment contrasts; typically, this is accomplished indirectly by differencing or by using residuals from neighboring plots to construct covariates. We examine an alternative approach in which the spatial heterogeneity is modeled directly. The model underlying our approach is similar to the model underlying a geostatistical kriging analysis and, as in the latter model, the observations are regarded collectively as a partial realization of a random field. A randomization study of uniformity trial data suggests that the random field approach often provides more accurate estimates of treatment contrasts than nearest-neighbor approaches. In addition, the random field approach is devoid of ambiguities as to the handling of border plots and is generally more flexible than nearest-neighbor approaches.

Process modeling of forest growth responses to environmental stress

Abstract: A survey of the available methods for evaluating the impact of atmospheric pollutants and other environmental stresses on forest growth, emphasizing quantitative means for predicting future growth and health in response to stress.

Procedures for two-sample comparisons with multiple endpoints controlling the experimentwise error rate.

Abstract: Clinical trials are often concerned with the comparison of two treatment groups with multiple endpoints. As alternatives to the commonly used methods, the T2 test and the Bonferroni method, O'Brien (1984, Biometrics 40, 1079-1087) proposes tests based on statistics that are simple or weighted sums of the single endpoints. This approach turns out to be powerful if all treatment differences are in the same direction [compare Pocock, Geller, and Tsiatis (1987, Biometrics 43, 487-498)]. The disadvantage of these multivariate methods is that they are suitable only for demonstrating a global difference, whereas the clinician is further interested in which specific endpoints or sets of endpoints actually caused this difference. It is shown here that all tests are suitable for the construction of a closed multiple test procedure where, after the rejection of the global hypothesis, all lower-dimensional marginal hypotheses and finally the single hypotheses are tested step by step. This procedure controls the experimentwise error rate. It is just as powerful as the multivariate test and, in addition, it is possible to detect significant differences between the endpoints or sets of endpoints.

On estimating HLA/disease association with application to a study of aplastic anemia.

Abstract: A multiplicative model is described relating HLA typing information to disease incidence. A likelihood-based method for estimating parameters in this model is proposed for use with data sets in which HLA haplotype information is available on a series of cases and their parents. This approach is extended to incorporate information from a matched control series for the purpose of estimating HLA and environmental risk factor effects simultaneously. The method is applied to data from aplastic anemia patients treated by bone marrow transplantation and the results are compared to unmatched case-control analyses using the same case series and several different control series.

Unequally spaced longitudinal data with AR(1) serial correlation.

Abstract: This paper discusses longitudinal data analysis when each subject is observed at different unequally spaced time points. Observations within subjects are assumed to be either uncorrelated or to have a continuous-time first-order autoregressive structure, possibly with observation error. The random coefficients are assumed to have an arbitrary between-subject covariance matrix. Covariates can be included in the fixed effects part of the model. Exact maximum likelihood estimates of the unknown parameters are computed using the Kalman filter to evaluate the likelihood, which is then maximized with a nonlinear optimization program. An example is presented where a large number of subjects are each observed at a small number of observation times. Hypothesis tests for selecting the best model are carried out using Wald's test on contrasts or likelihood ratio tests based on fitting full and restricted models.

Quantifying Precision of Mark-Recapture Estimates Using the Bootstrap and Related Methods

Abstract: SUMMARY Bootstrap techniques yield variance estimates under any model for which parameter estimates can be calculated, and are useful in cases where analytic variances are not available in closed form, or are available only if more restrictive assumptions are made. Here the application of bootstrap techniques to mark-recapture models is discussed. The approach also allows generation of robust confidence intervals, which extend beyond the permissible parameter range only if the mark-recapture model itself allows out-of-range parameter estimates. If an animal population is assumed to be closed (i.e., no death, birth, or migration), two further methods of obtaining confidence limits for population size are suggested. The first is based on a Robbins-Monro search for each limit, and the second applies the concept of a randomisation or permutation test. In the absence of nuisance parameters, both methods are exact apart from Monte Carlo variation and the limitations imposed by a discrete distribution. For the second, if all possible permutations are enumerated, Monte Carlo variation is eliminated.

The validity of inferences based on incomplete observations in disease state models.

Abstract: In many survival time studies or studies on the progression of a disease, information is often incomplete in the sense that it is known only that a patient has been in certain disease states at several time points. In this paper, conditions concerning the interrelationship between the disease process and the examination scheme (i.e., the pattern of examination times) are derived under which a valid statistical inference is possible. These conditions are confronted with examination schemes that are of practical importance in clinical research. A cancer marker study is used as an example to estimate the magnitude of the potential bias when the conditions derived are violated.