Multiple-endpoint models are widely used in drug development and other fields. It is common that the endpoints have different characteristics such as all continuous, all binary, or a mixture of them. This study investigates mixed responses, one continuous and one binary, correlated and observed simultaneously. It is an extension of our previous studies based on a bivariate probit model for two binary endpoints. We quantify the study goal with a utility function, construct locally two-stage D-optimal designs under the constraints, and use them as benchmarks for the two-stage designs with interim adjustment and fully adaptive designs. The simulation results suggest that the two-stage design is almost as efficient as the locally two-stage optimal design, as well as being logistically simpler than the fully adaptive design. We do not analyze asymptotic properties but confine ourselves to Monte-Carlo simulations to evaluate their properties for reasonable (practical) sample sizes.
Optimal Design Penalty Function Design Point Information Matrix Single Observation
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in to check access
Dragalin, V., V. Fedorov, and Y. Wu (2008). Two-stage design for dose-finding that accounts for both efficacy and safety. Statistics in Medicine27, 5156–5176.CrossRefGoogle Scholar
Fedorov, V. (1972). Theory of Optimal Experiments. New York: Academic Press.Google Scholar
Fedorov, V. and Y. Wu (2007). Dose finding designs for continuous responses and binary utility. Journal of Biopharmaceutical Statistics17, 1085–1096.CrossRefMathSciNetGoogle Scholar
Fedorov, V. V. and P. Hackl (1997). Model-oriented Design of Experiments. Berlin: Springer-Verlag.MATHGoogle Scholar
Lai, T. L. (2001). Sequential analysis: Some classical problems and new challenges. Statistica Sinica11, 303–351.MATHMathSciNetGoogle Scholar
Lai, T. L. and H. Robbins (1978). Adaptive design in regression and control. Proceedings of the National Academy of Sciences of the United States of America75, 586–587.MATHCrossRefMathSciNetGoogle Scholar
Rosenberger, W. F. and J. M. Hughes-Oliver (1999). Inference from a sequential design: Proof of a conjecture by Ford and Silvey. Statistics & Probability Letters44, 177–180.MATHCrossRefMathSciNetGoogle Scholar
Tate, R. (1955). The theory of correlation between two continuous variables when one is dichotomized. Biometrika42, 205–216.MATHMathSciNetGoogle Scholar