Latent transition analysis: Inference and estimation
Article first published online: 11 DEC 2007
Copyright © 2007 John Wiley & Sons, Ltd.
Statistics in Medicine
Volume 27, Issue 11, pages 1834–1854, 20 May 2008
How to Cite
Chung, H., Lanza, S. T. and Loken, E. (2008), Latent transition analysis: Inference and estimation. Statist. Med., 27: 1834–1854. doi: 10.1002/sim.3130
- Issue published online: 10 APR 2008
- Article first published online: 11 DEC 2007
- Manuscript Accepted: 4 OCT 2007
- Manuscript Received: 25 MAY 2007
- National Institute on Drug Abuse. Grant Numbers: 5-R03-DA021639, 1-P50-DA10075
- latent transition analysis;
- small samples;
- EM algorithm
Parameters for latent transition analysis (LTA) are easily estimated by maximum likelihood (ML) or Bayesian method via Markov chain Monte Carlo (MCMC). However, unusual features in the likelihood can cause difficulties in ML and Bayesian inference and estimation, especially with small samples. In this study we explore several problems in drawing inference for LTA in the context of a simulation study and a substance use example. We argue that when conventional ML and Bayesian estimates behave erratically, problems often may be alleviated with a small amount of prior input for LTA with small samples. This paper proposes a dynamic data-dependent prior for LTA with small samples and compares the performance of the estimation methods with the proposed prior in drawing inference. Copyright © 2007 John Wiley & Sons, Ltd.