Tao Liu PhD
PHP 2601: Linear and Generalized Linear Models
 

Course Info
Instructor: Tao Liu PhD
Location: 121 South Main St, Room 241
Time: 1:00 – 2:20 pm, Tuesday and Thursday
Office Hours: By appointment
Tel.: 401-863-6480

Course description (from Banner).
Generalized linear models (GLMs) provide a unifying framework for regression. Examples of GLM include linear regression, log-linear models, and logistic regression. GLMs can be used for modeling continuous, binary, ordinal, nominal, and count data. Topics of this course include model parameterization, parametric and semiparametric estimations, inferences, and model interpretations. Computing with standard software is introduced, with focus on epidemiology and clinical research applications.

Prerequisites.
Graduate level PHP2520 or PHP0257 or Undergraduate level APMA 1650 or APMA 0165.

Method of evaluation.
There will be 7-10 homeworks (30%) and three exams (exam I 20%; exam II 20%; and exam III 30%). The lowest grade of homeworks will be dropped. Students should work independently on homework, but discussion on general aspect of the course content is encouraged.

Text books
1. Seber, G.A. and Lee, A.J. (2003). Linear Regression Analysis (2nd Ed). John Wiley & Sons.
2. McCullagh, P. and Nelder, J.A. (1989). Generalized Linear Models (2nd Ed). Chapman & Hall.
3. (Optional) Agresti, A. (2002). Analysis of Categorical Data. New York: Wiley. (One copy is reserved at the CSS library)

Software. SAS, STATA and R. Some homework problems need moderate programming. Feel free to use the software of your preference.

PHP 2601 syllabus