Colloquium
- Home
- Undergraduate Programs
- Graduate Programs
- Courses
- Four-year Offerings
- WMU Calendar
- Final Exams
- People
- Statistical Consulting Center
- Newsletter
- Resources
- Careers in Statistics
- Contact Us
Statistics Colloquium
December 3 (Fri) 11 a.m.
Alavi Commons Room, 6625 Everett Tower
Derek Young
The Pennsylvania State University
Finite mixtures of regression models are appropriate when regression data are believed to belong to two or more distinct categories, but the categories themselves are unobserved. Mixtures of regressions are widely used in research areas like biomedical engineering, machine learning, economics, and psychology. However, one may desire greater flexibility in the modeling process when faced with more complex data structures. This has resulted in several recent generalizations of the standard fully parametric mixture of linear regressions model in the literature. In this talk, we present novel parametric and semiparametric extensions to the traditional mixture of linear regressions model along with a discussion of some estimation challenges for each model. We then present EM-based algorithms for estimating the parameters and summarize some numerical results. We also highlight the relevant R functions used for each analysis, which are available in the contributed package mixtools.
All statistics students are expected to attend.