Maximum Likelihood Estimation for Generalized Linear Models

Instructor(s):

  • Dean Lacy, Dartmouth College

This course introduces students to a number of useful statistical models that move beyond standard linear regression. Among the topics covered are logit and probit models for both binary and ordinal dependent variables, event count models, models for heteroskedastic regressions, and more. Maximum likelihood unifies these models by providing a single, coherent approach to estimation and a way of thinking about how data are generated. The mathematical background needed for the course is multiple regression using linear algebra. Attendance at the Mathematics for Social Scientists II lectures should prove useful.

Fees: Consult the fee structure.

Tags: logit, probit, nonlinear,

Course Sections

Section 1

Location: ICPSR -- Ann Arbor, MI

Date(s): June 23 - July 18

Time: 9:00 AM - 11:00 AM

Instructor(s):

  • Dean Lacy, Dartmouth College

Syllabus:

Found a problem? Use our Report Problem form to let us know.