The Sample Average Approximation Method for Estimating Latent Variable Models
Nuo Xi · OhioLink ETD Center (Ohio Library and Information Network) · 2008
This thesis investigates the usefulness of a simulation-based method, the socalled Sample Average Approximation (SAA) method, in estimating latent variable models for dichotomous data.The standard technique in dealing with latent variable models, i.e., the Marginal Maximum Likelihood Estimation (MMLE) method, requires calculating the marginal likelihood as the first step.But it is often difficult to First and foremost, I would like to thank my advisor Dr Michael Browne for allowing me to work on this thesis.Without his support, this work would not be possible.The time and the effort you have devoted in instructing me through the three-year time are much appreciated.I would also like to thank my other committee members, Dr Jay Myung, and Dr Michael Edwards, for agreeing to serve on my committee.You have always been quick to my questions, and I have received many valuable suggestions from you.