Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Update to -stjm- on SSC


From   "Crowther, Michael J." <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: Update to -stjm- on SSC
Date   Fri, 3 Feb 2012 10:52:29 +0000

Thanks to Kit Baum, a major update to -stjm- can be downloaded from SSC. This can be installed by typing -ssc install stjm, replace- in Stata.

stjm fits shared parameter joint models for longitudinal and survival data using maximum likelihood. A single continuous longitudinal response and a single survival outcome are allowed.  A linear mixed effects model is used for the longitudinal submodel, with fixed and/or random fractional polynomials of measurement time. Four choices are currently available for the survival submodel; the first being the flexible parametric survival model (see stpm2), modelled on the log cumulative hazard scale.  The remaining choices include the exponential, Weibull and Gompertz proportional hazard models. The association between the two processes can be induced via the default current value parameterisation, the first derivative of the longitudinal submodel (i.e. the slope), and/or a random coefficient such as the intercept. 

Updates include:

* Adaptive Gauss-Hermite quadrature has been implemented and is now the default technique to evaluate the joint likelihood, which requires far fewer quadrature points to achieve a high degree of accuracy compared to non-adaptive quadrature.
* All evaluator programs are substantially faster than the previous version of -stjm-
* -timeinteraction- option has been added allowing the user to interact covariates with time in the longitudinal sub-model
* -assoccovariates- option has been added allowing the user to specify covariates in the linear predictor of association parameters
* Predictions have been extended, including standard errors of the BLUP estimates. Survival sub-model predictions with option xb are now the average of the fixed portion of the model plus m draws from the random effects variance-covariance matrix.

A draft Stata Journal article describing the command can be found here:
http://leicester.academia.edu/MichaelCrowther/Papers/1369569/Joint_modelling_of_longitudinal_and_survival_data

Thanks,
Michael

Michael J. Crowther
Research Assistant in Medical Statistics
Centre for Biostatistics and Genetic Epidemiology
Department of Health Sciences
University of Leicester
Room 212, Adrian Building
University Road
Leicester LE1 7RH

E-mail: [email protected]
Phone: +44 (0)116 229 7278



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index