Conditional and Marginal Models: Another View

There has existed controversy about the use of marginal and conditional models, particularly in the analysis of data from longitudinal studies. We show that alleged differences in the behavior of parameters in so-called marginal and conditional models are based on a failure to compare like with like. In particular, these seemingly apparent differences are meaningless because they are mainly caused by preimposed unidentifiable constraints on the random effects in models. We discuss the advantages of conditional models over marginal models. We regard the conditional model as fundamental, from which marginal predictions can be made.

[1]  P. Thall,et al.  Some covariance models for longitudinal count data with overdispersion. , 1990, Biometrics.

[2]  L. Sheiner,et al.  Modelling of individual pharmacokinetics for computer-aided drug dosage. , 1972, Computers and biomedical research, an international journal.

[3]  F. Yates.  Reader reaction - regression-models for repeated measurements , 1982 .

[4]  L. Ryan,et al.  Analysis of dichotomous outcome data for community intervention studies , 2000, Statistical methods in medical research.

[5]  Thurston Dart,et al.  The Interpretation of Music , 1955 .

[6]  P. Meier,et al.  Choosing covariates in the analysis of clinical trials. , 1989, Controlled clinical trials.

[7]  Stephen Senn,et al.  The Many Modes of Meta , 2000 .

[8]  Jiming Jiang.  On maximum hierarchical likelihood estimators , 1999 .

[9]  Youngjo Lee,et al.  Analysis of ulcer data using hierarchical generalized linear models , 2002, Statistics in medicine.

[10]  Jiming Jiang.  REML estimation: asymptotic behavior and related topics , 1996 .

[11]  Youngjo Lee,et al.  HGLM versus conditional estimators for the analysis of clustered binary data , 2005, Statistics in medicine.

[12]  D. Cox,et al.  Parameter Orthogonality and Approximate Conditional Inference , 1987 .

[13]  P. Lahiri,et al.  A UNIFIED MEASURE OF UNCERTAINTY OF ESTIMATED BEST LINEAR UNBIASED PREDICTORS IN SMALL AREA ESTIMATION PROBLEMS , 2000 .

[14]  Jiming Jiang,et al.  Goodness-of-fit tests for mixed model diagnostics , 2001 .

[15]  J Norrie,et al.  Model inconsistency, illustrated by the Cox proportional hazards model. , 1995, Statistics in medicine.

[16]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[17]  David J. Spiegelhalter,et al.  Consensus and controversy in pharmaceutical statistics - Discussion , 2000 .

[18]  H. D. Patterson,et al.  Recovery of inter-block information when block sizes are unequal , 1971 .

[19]  M. Gail,et al.  Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates , 1984 .

[20]  Subject-by-Formulation Interaction in Bioequivalence: Conceptual and Statistical Issues , 2000, Pharmaceutical Research.

[21]  Zhenming Shun.  Another Look at the Salamander Mating Data: A Modified Laplace Approximation Approach , 1997 .

[22]  R. Schall.  Estimation in generalized linear models with random effects , 1991 .

[23]  N. Jewell,et al.  Some surprising results about covariate adjustment in logistic regression models , 1991 .

[24]  Jiming Jiang.  Wald consistency and the method of sieves in REML estimation , 1997 .

[25]  Youngjo Lee,et al.  Comparison of hierarchical and marginal likelihood estimators for binary outcomes , 2004, Comput. Stat. Data Anal..

[26]  A R Localio,et al.  Empirical Bayes Estimation of Random Effects Parameters in Mixed Effects Logistic Regression Models , 1999, Biometrics.

[27]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[28]  H. Goldstein.  Multilevel Statistical Models , 2006 .

[29]  S. Zeger,et al.  Longitudinal data analysis using generalized linear models , 1986 .

[30]  R. Stine,et al.  Non-parametric estimates of overlap. , 2001, Statistics in medicine.

[31]  P. W. Lane,et al.  Analysis of covariance and standardization as instances of prediction. , 1982, Biometrics.

[32]  J. Pearl,et al.  Confounding and Collapsibility in Causal Inference , 1999 .

[33]  Youngjo Lee,et al.  Hierarchical-Likelihood Approach for Mixed Linear Models with Censored Data , 2002, Lifetime data analysis.

[34]  J. Nelder,et al.  Double hierarchical generalized linear models , 2006 .

[35]  Youngjo Lee,et al.  Analyzing weather effects on airborne particulate matter with HGLM , 2003 .

[36]  John A. Nelder,et al.  Likelihood for random-effect models , 2005 .

[37]  N. Breslow,et al.  Bias correction in generalised linear mixed models with a single component of dispersion , 1995 .

[38]  J. Kalbfleisch,et al.  The effects of mixture distribution misspecification when fitting mixed-effects logistic models , 1992 .

[39]  J. Grizzle.  THE TWO-PERIOD CHANGE-OVER DESIGN AN ITS USE IN CLINICAL TRIALS. , 1965, Biometrics.

[40]  P. Heagerty,et al.  Misspecified maximum likelihood estimates and generalised linear mixed models , 2001 .

[41]  Peter McCullagh,et al.  Laplace Approximation of High Dimensional Integrals , 1995 .

[42]  P. Diggle,et al.  Analysis of Longitudinal Data , 2003 .

[43]  A. Garrett.  Therapeutic equivalence: fallacies and falsification , 2003, Statistics in medicine.

[44]  Maengseok Noh,et al.  HGLM modelling of dropout process using a frailty model , 2005, Comput. Stat..

[45]  D. Rom,et al.  Testing for individual and population equivalence based on the proportion of similar responses. , 1996, Statistics in medicine.

[46]  Y. Pawitan.  In all likelihood : statistical modelling and inference using likelihood , 2002 .

[47]  N. Breslow,et al.  Bias Correction in Generalized Linear Mixed Models with Multiple Components of Dispersion , 1996 .

[48]  Martin Crowder,et al.  On the use of a working correlation matrix in using generalised linear models for repeated measures , 1995 .

[49]  R. Fisher.  014: On the "Probable Error" of a Coefficient of Correlation Deduced from a Small Sample. , 1921 .

[50]  Youngjo Lee,et al.  Modelling and analysing correlated non-normal data , 2001 .

[51]  A. Welsh,et al.  ASYMPTOTIC PROPERTIES OF RESTRICTED MAXIMUM LIKELIHOOD (REML) ESTIMATES FOR HIERARCHICAL MIXED LINEAR MODELS , 1994 .

[52]  Stephen Senn,et al.  Statistical issues in bioequivalance , 2001 .

[53]  R. Hogg,et al.  Symmetrically distributed and unbiased estimators in linear models , 1982 .