Biostatistics Advance Access originally published online on July 11, 2007
Biostatistics 2008 9(2):249-262; doi:10.1093/biostatistics/kxm026
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A penalized latent class model for ordinal data
Department of Biostatistics, Harvard University, 655 Huntington Avenue, Boston, MA 02115, USA sdesanti{at}hsph.harvard.edu
Department of Pathology, CNY-7015, Massachusetts General Hospital, 149, 13th Street, Charlestown, MA 02129, USA
Department of Biostatistics, Harvard University, 655 Huntington Avenue, Boston, MA 02115, USA
* To whom correspondence should be addressed.
Latent class models provide a useful framework for clustering observations based on several features. Application of latent class methodology to correlated, high-dimensional ordinal data poses many challenges. Unconstrained analyses may not result in an estimable model. Thus, information contained in ordinal variables may not be fully exploited by researchers. We develop a penalized latent class model to facilitate analysis of high-dimensional ordinal data. By stabilizing maximum likelihood estimation, we are able to fit an ordinal latent class model that would otherwise not be identifiable without application of strict constraints. We illustrate our methodology in a study of schwannoma, a peripheral nerve sheath tumor, that included 3 clinical subtypes and 23 ordinal histological measures.
Received July 10, 2006; revised December 18, 2006; revised April 16, 2007; accepted for publication May 9, 2007.