This course is discontinued

BIO9290 – Latent variable modeling

Schedule, syllabus and examination date

Choose semester

Course content

  • Multilevel, longitudinal, and structural equation models.
  • Generalized mixed models, random coefficient models, item response models, factor models, panel models, latent class models and frailty models.
  • Identification and equivalence.
  • Estimation and inference based on the likelihood function
  • Automatic differentiation
  • Laplace’s method for integrating the likelihood
  • Model selection
  • Applications
  • Software: R-nlme, ADMB-RE

Learning outcome

Latent variable models are used in most empirical disciplines, but under various names and with various motivations and methods of analysis. The aim is to establish a unified conceptual framework for modeling and analysis. Various examples or applications from genetics, ecology, evolution, economics and other fields will be reviewed for the purpose of cross-fertilization. The course is aimed at doctoral students from genetics, evolution and ecology, as well as from economics and other social sciences– in the hope that communality in modeling needs and analytical approach provides a fertile ground for exchanging ideas. The free ware R can handle more standard mixed models (models with both manifest (fixed) and latent (random) components). Bigger and more complex models are better handled by the commercial system ADMB-RE, which is available for free for a limited period. The RE-add on to handle random effect models is developed by Hans Julius Skaug and David Fournier.

Admission

PhD candidates from the University of Oslo should apply for classes and register for examinations through Studentweb.

If a course has limited intake capacity, priority will be given to PhD candidates who follow an individual education plan where this particular course is included. Some national researchers’ schools may have specific rules for ranking applicants for courses with limited intake capacity.

PhD candidates who have been admitted to another higher education institution must apply for a position as a visiting student within a given deadline.

Prerequisites

Formal prerequisite knowledge

To participate in this course, you have to have admission to a Ph.d.-Programme.

Recommended previous knowledge

Enough mathematics, statistics or econometrics to read and appreciate the book by Skrondal and Rabe-Hesket. Take a look!

Teaching

The course will be conducted in three intensive pulses (week 5, 7 and 9), with two hours of lectures/seminars/computer sessions five days each of these three weeks. Students (and other participants, such as Thomas Hansen) are invited to present their own data and problems.

Curriculum

Most of Skrondal, A. and Rabe-Hesketh, S. 2004. Generalized latent variable modeling. Chapman & Hall.

Examination

There will be one compulsory written home assignment, a group-presentation of papers after the course, and a take-home exam. Passing the home assignment and the group-presentation is a prerequisite for being allowed to take the take-home exam. Pass/fail.

Facts about this course

Credits

10

Level

PhD

Teaching

Spring 2006

Examination

Spring 2006

Teaching language

English