Course
Generalized Linear Models (STA600)
THIS COURSE WILL NOT BE OFFERED SPRING 2014.
Introduction to glm, which is a generalization of (multiple) regression for normally distributed responses to responses from a larger class of distributions, especially discrete responses. Theory for glm’s with application to regression for normally distributed data, logistic regression for binary and multinomial data; Poisson regression and survival analysis. Applications to data, principles of statistical modeling, estimation and inference are emphasized. Likelihood theory.
Dette er emnebeskrivelsen for studieåret 2013-2014. Merk at det kan komme endringer.
Semesters
Fakta
Emnekode
STA600
Vekting (stp)
10
Semester undervisningsstart
Spring
Undervisningsspråk
English
Antall semestre
1
Vurderingssemester
Spring
Content
Learning outcome
After having completed the course one the student should:
• Know the main theory for Generalized Linear Models (glm);
• Know how regression with binary, multinomial, Poisson- and survival time responses may be done;
• Understand use of likelihood estimation generally and especially for generalized linear models, and
• Be able to use glm’s in practical use of real data.
Forkunnskapskrav
Mathematical Methods 1 (MAT100)
Mathematical Methods 2 (MAT200)
Probability and Statistics 1 (STA100)
Probability and Statistics 2 (STA500)
Anbefalte forkunnskaper
Exam
Fagperson(er)
Head of Department:
Bjørn Henrik AuestadCourse coordinator:
Arild BulandCourse coordinator:
Tore Selland KleppeCourse coordinator:
Jörn SchulzMethod of work
THIS COURSE WILL NOT BE OFFERED SPRING 2014.
4 hours lectures and 2 hours problem solving per week.
Åpent for
Emneevaluering
Litteratur
Annette J. Dobson: An introduction to generalized linear models, second ed.
Ch.: 1 – 11.
Alternative books will be considered. If another book is chosen information about this will be given at the start of the term.