CMStatistics 2016: Start Registration
View Submission - CMStatistics
B0630
Title: Computational construction of Minimax designs in binary response and heteroscedastic simple linear regression models Authors:  Jesus Lopez-Fidalgo - University of Navarra (Spain) [presenting]
Victor Casero-Alonso - University of Castilla-La Mancha (Spain)
Abstract: Binary response models are used in many real applications. For these models the Fisher Information Matrix (FIM) is proportional to the FIM of a weighted simple linear regression model. The same is also true when the weight function has a finite integral. Thus, optimal designs for one binary model are also optimal for the corresponding weighted linear regression model. The main objective is to provide a tool for the construction of MV-optimal designs, minimizing the maximum of the variances of the estimates, for a general design space. MV-optimality is a potentially difficult criterion because of its non-diferentiability at equal variance designs. A methodology for obtaining MV-optimal designs where the design space is a compact interval $[a,b]$ will be given for several standard weight functions. This will allow us to build a user-friendly computer tool based on Mathematica to compute MV-optimal designs. Users can know the type of the optimal design and the exact support points and design weights. Some illustrative examples will show a representation of MV-optimal designs in the Euclidean plane taking $a$ and $b$ as the axes. The applet provided will be explained using two relevant models for a weighted linear regression model and for a binary response model.