Quantal dose-response experiments are routinely conducted to examine the response rates at various dose levels of interest. Because of side effects, lack of efficacy or other problems, not all responses are always observed in practice. Consequently, our estimate can be inefficient. In this work, we incorporate information on the expected pattern of potentially missing observations in the study and construct multiple-objective optimal designs for dose response studies that account for the potential missing observations at the onset. The performances of designs with and without accounting for potential missing trials are compared in terms of the precision of related parameter estimation. |