Department of Statistics, The Ohio State University
Statistics and Biostatistics Colloquium Series

This talks describes a method for generating split designs that are efficient for estimating the fixed effects for a design with a split plot structure. One advantage of the method is that it does not require the prior specification of a candidate set. This makes the production of split designs computationally feasible in situations where the candidate set is too large to be tractable.

The method supports creation of designs for any number of whole plots greater than the minimum required to fit the whole plot fixed effects. The total sample size can be any number greater than than the number of whole plots. The factors can be continuous or categorical with arbitrary numbers of levels. The model can be any linear regression model consistent with the number of whole plots and sample size. It can include arbitrary polynomial terms in the continuous factors and interactions to any order in the categorical factors.

This is joint work with Peter Goos.

Meet the speaker in Room 212 Cockins Hall at 4:30 p.m. Refreshments will be served.