Titlebook: Biostatistics; Advances in Statisca Ian B. MacNeill,Gary J. Umphrey,V. Krishna Jandhya Book 1987 D. Reidel Publishing Company, Dordrecht, H

Musket

Estimation Under the Correlated Logistic Model,lculation of maximum likelihood estimates under this model is costly, alternative estimators based on the usual logistic model, with or without dummy varibles, are asymptotically biased (in general) and yield standard errors that are incorrect. An adjustment for the standard error of the usual logis

adumbrate

Inferences about Interclass and Intraclass Correlations from Familial Data,ood ratio test is proposed for testing the equality of intraclass correlations in two multivariate normal populations. The asymptotic null distribution of the proposed test statistic is obtained. A test procedure based on Fisher’s .-transformation is also discussed.

absorbed

Maximum Likelihood Estimation in Quantal Response Bioassay,obtained using the Fieller method. A procedure is presented in this paper based on transformations of the parameters of interest, designed to facilitate normal, ., or log . approximations to the observed maximized or profile likelihood function. Comparisons with the Fieller method and applications t

debunk

Learning Regression Analysis by Simulationo the diagnosis and treatment of patients with various forms of cancer (e.g., Dembo ., 1982; Ciampi ., 1981) or coronary heart disease (e.g., Mock .., 1982), or those related to prospective studies, for example on the risk of coronary heart disease (Keys .., 1972). Such data bases typically contain

貞潔

本土

Peculate

Zhengming Ding,Handong Zhao,Yun Fuoassay. Previously proposed tests which incorporate historical control information assume at least tacitly that the concurrent and historical control data arise from the same distribution. In this paper, we introduce a twostage procedure which utilizes historical data only if the response rate in th

Glossy

進(jìn)取心

Wei-Po Lee,John Hallam,Henrik Hautop Lundd provide adequate approximations provided the parameters . and . of the binomial model are such that . is not small. Although the asymptotic expansions are algebraically complicated, they have been mathematically expressed in such a way that general coefficients can be isolated and calculated.

evaculate

Wei-Po Lee,John Hallam,Henrik Hautop Lund common odds ratio from multiple 2 × 2 tables are reviewed in this paper, with particular emphasis on the point estimators, their asymptotic properties, and what is known about finite-sample properties. Based on research to date, the conditional maximum likelihood and Mantel-Haenszel estimators are