-- KathrynStoltzfus - 27 Feb 2018

Nature-Inspired metaheuristic algorithms for finding optimal designs for complex biostatistical problems

Algorithms are practical ways to find optimal experimental designs. Most published work in the statistical literature concern optimal design problems for relatively simple models with a small number of factors under a user-specified criterion. With big data, there are increasingly high-dimensional design problems with many factors and current algorithms do not work well. Nature-inspired metaheuristic algorithms are general and powerful optimization tools that seem to be under-utilized in statistical research. I present an overview of such algorithms, describe their numerous advantages over current algorithms for finding efficient designs, and demonstrate how they quickly find single or multiple-objective optimal designs for dose response studies and for tackling high-dimensional optimal design problems in biostatistics.
Topic revision: r1 - 27 Feb 2018, KathrynStoltzfus
 

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback