PhD Qualifying Examination


"A Survey of Latent Structure Models"

By Mr. Yi Wang


Abstract:

Latent structure models are popular tools for analyzing discrete data in
social and behavioral sciences. These models are based on two basic
assumptions: (1) the existence of an unobserved or latent variable; (2)
all observed or manifest variables are independent from each other given
the state of the latent variable. The second assumption is often called
local independence. These assumptions make latent structure models
capable of revealing hidden structures among observed variables and
investigating unobservable concepts. In this survey, we first review
existing latent structure models. We then discuss how to deal with
situations when the assumption of local independence is violated.


Date:     		Monday, 17 January 2005

Time:                   2:00p.m.-4:00p.m.

Venue:                  Room 4502
			lifts 25-26

Committee Members:      Dr. Nevin Zhang (Supervisor)
                        Dr. D.Y. Yeung (Chairperson)
			Dr. Albert Chung
			Dr. James Kwok


**** ALL are Welcome ****