We suggest a procedure for
finding approximate local D-optimal designs for GLM. However, an alternative
algorithm is proposed in the one-stage designs section,
that is both more efficient and more ordered. Interest in the approximate
algorithm may be found for two reasons:
1.
It gives
an intuitive understanding of the expected locations which are D-optimal for a
binary response; and
2.
It may
be preferred for designs of very high dimension.
Publication:
Hovav A. Dror and David M. Steinberg (2005), Approximate Local D-optimal Experimental Design for Binary Response, Technical Report RP-SOR-0501, Tel Aviv University
Back to the parent page: Experimental Design for GLM