I-Biostat Software Longitudinal and Incomplete Data

Software - Longitudinal and Incomplete Data

This section contains models in the area of Longitudinal and Incomplete Data.
For software in other areas, go to the general overview of software.

For Longitudinal and Incomplete Data, the following software is available:

A SAS macro for linear mixed models with finite normal mixtures as random-effects distribution
A SAS macro for local influence analysis in linear mixed models
A SAS macro for nonlinear and generalised linear mixed models with finite normal mixtures as random-effects distribution
A SAS macro for fitting a multivariate linear mixed model using the pairwise approach
A SAS macro for fitting a multivariate generalised linear mixed model using the pairwise approach
Diagnostic tools for misspecification in generalized linear mixed models
Analysis and sensitivity analysis for incomplete longitudinal data
Programs for use of identification restrictions strategies to handle pattern mixture models

A SAS MACRO FOR LINEAR MIXED MODELS WITH FINITE NORMAL MIXTURES AS RANDOM-EFFECTS DISTRIBUTION
Program: hetmixed	Version: 1.1	Software: SAS	Version: 6.12	Last updated: 12/03/2005
Prepared by:	Arnost Komárek, Geert Verbeke
Description:	A description of the macro in pdf-format and the SAS script files used for the reported examples can be downloaded together with the macro.
download (size 123KB)
top

A SAS MACRO FOR LOCAL INFLUENCE ANALYSIS IN LINEAR MIXED MODELS
Program: locinfl	Version: 1	Software: SAS	Version: 6.12 or 8.2	Last updated: 12/03/2005
Prepared by:	Franz Torres Barbosa (Center for Molecular Immunology, Habana, Cuba) and Geert Verbeke
Description:	The macro performs a local influence analysis for the detection of influential subjects in linear mixed models. We refer to Verbeke and Molenberghs (Section 3.13.2) for a description of the procedure, and to Lesaffre and Verbeke for the technical details. Three examples of how to use the macro are also given. References: Verbeke G. and Molenberghs G. (1997), 'Linear Mixed Models in Practice: A SAS-Oriented Approach', Lecture Notes in Statistics 126, New-York: Springer-Verlag. Lesaffre E., Verbeke G. (1997) 'Local influence in linear mixed models', Biometrics, 54, 570-582.
download (size 123KB)
top

A SAS MACRO FOR NONLINEAR AND GENERALISED LINEAR MIXED MODELS WITH FINITE NORMAL MIXTURES AS RANDOM-EFFECTS DISTRIBUTION
Program: hetnlmixed	Version: 1.2.5	Software: SAS	Version: 6.12 or 8.2	Last updated: 12/03/2005
Prepared by:	Bart Spiessens, Geert Verbeke, Arnost Komárek, Steffen Fieuws
Description:	A paper in pdf-format describing the macro can be downloaded together with the macro.
download (size 123KB)
top

A SAS MACRO FOR FITTING A MULTIVARIATE LINEAR MIXED MODEL USING THE PAIRWISE APPROACH
Program: jointpairv1.sas	Version: 1	Software: SAS	Version: 9.1.3	Last updated: 13/12/2006
Prepared by:	Steffen Fieuws
Description:	A SAS macro using the pairwise approach to fit a multivariate linear mixed model. References: Fieuws S and Verbeke G. Pairwise Fitting of Mixed Models for the Joint Modelling of Multivariate Longitudinal Profiles. Biometrics, 2006, 62 (2), 424-431. Abstract
download (size 30KB)
top

A SAS MACRO FOR FITTING A MULTIVARIATE GENERALISED LINEAR MIXED MODEL USING THE PAIRWISE APPROACH
Program: MGLMM_modelA.sas	Version: 1	Software: SAS	Version: 9.1.3	Last updated: 11/12/2006
Prepared by:	Steffen Fieuws
Description:	Use of the pairwise approach to fit a multivariate generalised linear mixed model. More specifically, the SAS code (and data) are given to fit model A discussed on page 453 of Fieuws S., Verbeke, G., Boen, F. and Delecluse, C. (2006) "High dimensional multivariate mixed models for binary questionnaire data", Applied Statistics, 55 (4), 449-460. Abstract
download (size 14KB)
top

DIAGNOSTIC TOOLS FOR MISSPECIFICATION IN GENERALIZED LINEAR MIXED MODELS
Program: TESTNLMIXED	Version: 1	Software: SAS	Version: 9.1.3	Last updated: 26/03/2008
Prepared by:	Saskia Litière
Description:	a macro implementing diagnostic tools for misspecification in GLMM, as described in Alonso, Litière, Molenberghs (2008), 'A Family of Tests to Detect Misspecifications in the Random-effects Structure of Generalized Linear Mixed Models ', Computational Statistics & Data analysis, doi:10.1016/j.csda.2008.02.033., Litière, Alonso, Molenberghs (2008), 'Testing for Misspecification in Generalized Linear Mixed Models', Submitted for publication.
download (size 10KB)
top

ANALYSIS AND SENSITIVITY ANALYSIS FOR INCOMPLETE LONGITUDINAL DATA
Program: MCAR	Version:	Software: SAS	Version:	Last updated:
Prepared by:
Description:	Methods to deal with incomplete longitudinal data under the missingness mechanisms MCAR, MAR, and MNAR: (1) MCAR analyses : LOCF & CC (using SAS) (2) MAR analyses : direct-likelihood, Weighted GEE, and multiple imputation combined with GEE (using SAS) (3) MNAR & sensitivity analyses: - Diggle-Kenward selection modelling and local influence (using SAS) - Pattern-mixture models (using SAS) - Shared parameter models & Latent-class mixture models (extension of SPM) (using GAUSS)
download (size 667KB)
top

PROGRAMS FOR USE OF IDENTIFICATION RESTRICTIONS STRATEGIES TO HANDLE PATTERN MIXTURE MODELS
Program: pmmstrat	Version:	Software:	Version:	Last updated:
Prepared by:	Herbert Thijs
Description:
download (size 126KB)
top