-
作者：PATEL, HI
摘要：A multivariate model is proposed for analysing repeated measures designs with changing or within-subject covariates. Under the assumption of multivariate normality the likelihood ratio tests are derived for testing certain linear hypotheses. Some special designs are discussed and a numerical example is given to illustrate the method for analysing a two-period crossover design.
-
作者：PORTIER, CJ
摘要：In most survival-sacrifice experiments for the detection and quantification of risks from chronic exposure to chemical agents, the onset of the condition of interest is not clinically observable. Cancer is typically such an event. In this paper, an approximate maximum likelihood method is proposed for parametric estimation of the distribution of unobservable tumour onset times in the presence of competing risks, when cause-of-death information is unavailable.
-
作者：HANNAN, EJ; KAVALIERIS, L; MACKISACK, M
-
作者：TSIATIS, AA; ROSNER, GL; TRITCHLER, DL
作者单位：University of Toronto; University Health Network Toronto; Harvard University; Harvard University Medical Affiliates; Dana-Farber Cancer Institute
-
作者：TAYLOR, JMG
作者单位：University of California System; University of California Los Angeles
摘要：Power transformations for achieving distributional symmetry are discussed. Estimates of the transformation power are based on general measures of symmetry. They are consistent and asymtotically normal. Use of the skewness coefficient as a measure of symmetry is optimal in an important special case. The methods are compared to the likelihood methods of Box et Cox (1964) and alternative methods of Hinkley. The Box-Cox method or a robust adaptation of it (Carroll, 1980; Bickel et Doksum, 1981) is...
-
作者：MANTEL, N
-
作者：GREEN, PJ
摘要：Spatial methods for the analysis of agricultural field experiments are represented here as smoothing methods applied simultaneously with the estimation of treatment effects. Selection of both the form of the smoother and the degree of smoothing required may be based on cross-validation. Particular emphasis is placed in this paper on generalized least squares estimation in linear models, but the principle applies quite generally.
-
作者：REID, N; CREPEAU, H
作者单位：University of British Columbia
摘要：Influence functions for the regression parameters in the proportional hazards model are presented. Empirical influence functions, computed for each observation and each covariate, can be useful in an informal way to identify influential observations. This is illustrated on the Stanford heart transplant data and 2 other examples.
-
作者：ZEGER, SL; LIANG, KY; SELF, SG
作者单位：Johns Hopkins University
摘要：Extensions of logistic regression to the case where the binary outcome variable is observed repeatedly for each subject are described. Two working models are proposed that lead to consistent estimates of the regression parameters and of their variances under mild assumptions about the time dependence within each subject''s data. The efficiency of the proposed estimators is examined. An analysis of stress in mothers with infants is presented to illustrate the proposed method.
-
作者：ANSLEY, CF; KOHN, R
