General

Course prerequisites: ST 512, ST 552

General Education Requirements: Not applicable.

Student learning outcomes for the course: The student will be familiar with general nonlinear parametric statistical models for univariate and multivariate continuous and discrete response, and with using standard statistical software for their analysis.

Required Textbooks: None. The course will be taught from Professor Davidian's class notes.

Course organization and scope:

1. Introduction and motivation
2. Models for univariate response
   1. Introduction to nonlinear models
   2. Implementation of generalized least squares (GLS), iteratively reweighted least squares
   3. Generalized (non)linear models, quasilikelihood
   4. Normal theory maximum likelihood (ML)
   5. Unknown parameters in the variance function
   6. Detecting and modeling nonconstant variance
   7. Large sample theory - a casual approach o The "folklore" theorem and "optimality" of GLS
   8. Linear vs. quadratic estimating equations for the regression parameter
   9. Effect of estimating weights in GLS
   10. Estimation of unknown parameters in variance function models
3. Models for multivariate response
   1. Modeling multivariate response - sources of correlation and "subject-specific" vs. "population-averaged" approaches
   2. Generalized estimating equation methods for population-averaged models
   3. Nonlinear and generalized linear mixed effects (subject-specific) models - approximate and "exact" methods

Reading assignments: see the reading assignments page.

Homework assignments: see the homework assignments page.

Tests: You will take three take-home tests. Each test will cover parts of the topics 2 and 3 listed above. Test questions will be similar to exercises on the preceding homework assignments. See the test schedule page.

Grades: The course grade will be based on homework (10%), and tests (30% each). The homework grade will reflect completeness and timeliness, not correctness; you are encouraged to work together to master these exercises. Tests are of course graded for correctness, and must be your own individual work.

Course grades will be rounded to the nearest %, and letter grades will be assigned on the basis A+ = 98-100, A = 93-97, A- = 90-92, B+ = 87-89, B = 83-87, and so on.

Auditors are excused from the tests, but meet all other course requirements, including timely submission of homework.

Incomplete grades and late assignments: An incomplete grade will be given only in accordance with NCSU grading policy. Homework will be accepted late only with prior approval from the instructor.

Attendance, absences, and makeup work: Full participation in classes and examinations is expected of all students. Students are expected to inform the instructor in advance about any anticipated excused absences. See the university attendance regulation for university definitions of excused absences.

Academic Integrity:

1. NCSU has a policy on academic integrity, which you may find in the Code of Student Conduct.

2. It is the understanding and expectation of the instructor that the student's signature on any test or examination means that the student neither gave nor received unauthorized aid.

3. As noted above, you are encouraged to work together to master the exercises assigned as homework. All tests and the final examination are submitted for a grade, and must be your own individual work.

Students with disabilities: Reasonable accommodations will be made for students with verifiable disabilities. In order to take advantage of available accommodations, students must register with Disability Services for Students at 1900 Student Health Center, Campus Box 7509, 515-7653. http://www.ncsu.edu/provost/offices/affirm_action/dss/ . For more information on NC State's policy on working with students with disabilities, please see the Academic Accommodations for Students with Disabilities Regulation.

Laboratory safety and field trips: Not applicable.

Statement on extra expenses: Not applicable.

Statement on transportation: Not applicable.