When: 17 mars 2021, kl. 13-14.
Where: This seminar is given online. E-mail Dan Hedlin if you want to attend.


Much of traditional optimal design theory relies on specifying a model with only a small number of parameters. In many applications, such models will give reasonable approximations. However, they will often be found not to be fully correct when enough data are at hand. We consider a low-dimensional model with a distortion term. Our objective is to estimate the combined model, including the distortion term, based on data collected in a clinical trial. In our situation, the low-dimensional regression model is a fixed effect and a Brownian bridge as the distortion term is a random effect in a mixed-effects model. Since we are interested in estimating the combination of fixed and random effect, our aim is to predict within the mixed model. After constructing the Best Linear Unbiased Estimator and Predictor in our model, we describe how we minimize the predictor error using an optimal design. Many algorithms can be used in order to construct an optimal design. We apply here Fedorov’s algorithm which exchanges observations between the design points. By performing the algorithm built on the distorted model, we present the optimal design in different cases.