Optimal Experimental Design7FIDA09, 2021VT
|
|
Course plan
The course is about optimal experimental design – planning of experiments. The basic idea of design optimization (best estimation of unknown model parameters, information or moment matrix, etc.) and commonly used design criteria in linear models are the main parts of the course. Besides optimal designs in classical linear models, optimal designs for estimation and prediction of fixed and random effects in particular mixed models will be discussed.
Contents (tentative)
1. Introduction
2. Optimal Design in Fixed Effects Models
2.1. Best linear unbiased estimator
2.2. Information Matrix
2.3. Design Criteria
2.4. Examples
3. Optimal Design in Random Coefficients Regression Models
3.1. Models with Known Population Parameters
3.1.1. Best Estimation
3.1.2. Design Criteria
3.1.3. Examples
3.2. Models with Unknown Population Parameters
3.2.1. BLUE and BLUP
3.2.2. Information Matrices
3.2.3. Design Criteria
3.2.4. Examples
4. Discussion on Optimal Designs in More Complex Models
Literature
Atkinson, A. C. and Donev, A. N. (1992). Optimum Experimental Designs. Oxford
University Press, Oxford.
Fedorov, V. (1972). Theory of Optimal Experiments. Academic Press, New York.
Fedorov, V. and Leonov, S. (2013). Optimal Design for Nonlinear Response
Models. CRC Press, Boca Raton.
Pukelsheim, F. (1993). Optimal Design of Experiments. Wiley, New York.
Silvey, S. D. (1980). Optimal Design. Chapman & Hall, London.
Examination and HEC
3 credits
To pass the examination the participants are expected to solve the seminar
exercises, to send their solutions to the teacher before the deadlines and to
be able to present the solutions in the seminars. (There will be no written
exam.)
Lecturer:
Maryna Prus
Page responsible: Director of Graduate Studies