Optimal Designs for Factorial Experiments with Binary Response

Talk announcement:
HIIT Seminar Kumpula, Friday May 27 10:15, Exactum C222

Speaker:
Abhyuday Mandal
University of Georgia, USA

Title:
Optimal Designs for Factorial Experiments with Binary Response

Abstract: 
We consider the problem of obtaining locally D-optimal designs for
factorial experiments with qualitative factors at two levels each with
binary response and identify optimal allocation of runs for $2^k$ and
$2^{n-k}$ designs. For the $2^2$ factorial experiment with main effects
model we obtain optimal designs analytically in special cases and
demonstrate how to obtain a solution in the general case using cylindrical
algebraic decomposition. The optimal designs are shown to be robust to the
choice of the assumed values of the prior and when there is no basis to
make an informed choice of the assumed values we recommend the use of the
uniform design, i.e., the design that assigns equal number of observations to 
each of the four points. We develop efficient numerical techniques for solving
this very high dimensional optimization problem. For the general $2^k$ case
we show that the uniform design has a maximin property. We also identify
the Bayesian D-optimal designs and compare their performances with locally
D-optimal designs.

Biography: 
Abhyuday Mandal is an assistant professor in the University of
Georgia, USA. He got his Bachelor’s and Master’s degrees in Statistics
from Indian Statistical Institute in Calcutta, India. Then he got another
Master’s degree in Statistics from University of Michigan and Ph.D. degree 
in 2005 from Georgia Institute of Technology, USA. His research interests
are Design of experiments, optimization techniques and genetic algorithms
and fMRI data analysis.


Welcome!
--Matti Järvisalo