ST705 - Linear Models and Variance Components
- Prerequisites: Corequisite: ST 702
- Term & Frequency: Spring
- Student Audience: PhD students in Statistics and other graduate students needing a thorough understanding of the theory of linear models.
- Credit: 3 credits
- Recent Texts: John F. Monahan, A Primer on Linear Models, Chapman & Hall/CRC Press, 2008
- Recent Instructors: L. Stefanski, Arnab Maity
- Background and Goals: The course covers the theory underlying linear statistical models, and provides the necessary theoretical foundation for understanding many advanced statistical methods and for doing methodological research in statistics.
- Content: General linear model; review of linear algebra; generalized inverses; solving linear equations; projections; linear least squares and the normal equations; estimability; Gauss-Markov Theorem; generalized least squares; multivariate normal distribution; central and non-central Chi-squared and F distributions; distributions of quadratic forms; general linear hypothesis; linear models with random effects; variance components;
- Alternatives: ST 503 (for students requiring only familiarity with linear model methodology)
- Subsequent Courses: ST 793, all other advanced courses in Statistics
S1 2017 Sections:
|-||/ - |