Optimal Crossover Designs for Generalized Linear Models

TL;DR

This paper develops locally D-optimal crossover designs for generalized linear models, accounting for correlation structures among repeated measures, and demonstrates their robustness and efficiency through simulations and real data examples.

Contribution

It introduces a method to identify optimal crossover designs under generalized linear models with various correlation structures, enhancing design efficiency.

Findings

Optimal allocations are robust to correlation assumptions.

Two-stage designs outperform uniform designs in correlated responses.

Real data example illustrates practical application and benefits.

Abstract

We identify locally $D$-optimal crossover designs for generalized linear models. We use generalized estimating equations to estimate the model parameters along with their variances. To capture the dependency among the observations coming from the same subject, we propose six different correlation structures. We identify the optimal allocations of units for different sequences of treatments. For two-treatment crossover designs, we show via simulations that the optimal allocations are reasonably robust to different choices of the correlation structures. We discuss a real example of multiple treatment crossover experiments using Latin square designs. Using a simulation study, we show that a two-stage design with our locally $D$-optimal design at the second stage is more efficient than the uniform design, especially when the responses from the same subject are correlated.