Three-Toed Sloth

Next Week at the Statistics Seminar: "Methods and Models for Time-Dependent Relational Data"

Next week at the CMU statistics seminar:

Andrew C. Thomas, "Methods and Models for Time-Dependent Relational Data"

Abstract: In recent years there has been a proliferation of relational data, including studies of social networks, where outcomes are measured over time. Methods that use binary outcomes traditionally rely on Markov chain assumptions that indicate the generation, maintenance and severing of network ties, often relying on dichotomized perceptions of relationships. I present a hierarchical latent variable model for time-dependent relations that makes use of dynamic programming techniques for its solution, and discuss the expansion of this model to non-binary ties, with several real and synthetic examples of social and professional interactions.

Date and place: Monday, 6 April 2009, 4-5 pm in Porter Hall 125C

Free and open to the public.

Networks; Enigmas of Chance