A Maximin Φp-Efficient Design for Multivariate Generalized Linear Models

Experimental designs for a generalized linear model (GLM) often depend on the specification of model, including link function, predictors, and unknown parameters, such as regression coefficients. To deal with uncertainties these specifications, it is important to construct optimal high efficiency under uncertainties. Existing methods Bayesian experimental use prior distributions specifications incorporate into design criterion. Alternatively, one can obtain by optimizing worst-case respect specifications. In this work, we propose new Maximin $\Phi_p$-Efficient (or Mm-$\Phi_p$ short) which aims at maximizing minimum $\Phi_p$-efficiency Based theoretical properties proposed criterion, develop an efficient algorithm sound convergence design. The performance assessed through several numerical examples.