STAT 591 - Fall 2009
Statistical Modeling with Latent Variables
Instructor: Elena Erosheva
|Latent variables are theoretical constructs that can not be
observed or measured directly. Such variables arise mainly, but not exclusively,
in the social sciences. The earliest example goes back to Spearman's 1904
one-factor general intelligence model. Modern statistical methodology is
abundant with latent variable approaches that include mixture models, potential
outcome models, latent class models, item response theory models, factor analysis
and group-based trajectory models, to name a few.
In this course, we will start by reviewing a general framework for latent variable modeling. We will consider different types of likelihood that arise in this framework and corresponding estimation strategies. The course will cover a series of readings starting with articles on such classic examples as latent class and the Rasch models. We will then move on to other types of latent variable modeling, including mixed membership, causal models with latent variables, random effects growth and group-based trajectory models.
We will meet for two hours every week. The format will be that of a reading group: each time participants will prepare and lead the discussion of a paper or a book chapter, or a few related publications. Credit will be given for participation.
|If you would like to request academic accommodations due to a disability, please contact Disabled Student Services, 448 Schmitz, 543-8924 (V/TTY). If you have a letter from Disabled Student Services indicating you have a disability that requires academic accommodations, please present the letter to me so we can discuss the accommodations you might need for this class.|