Study Designs for the Estimation of the Hill Parameter in Sigmoidal Response Models
Sigmoidal models are frequently considered for the description of the dose-response relationship in dose-finding studies. Designs for these models depend on unknown parameters, which enter the model in a nonlinear way. A particular problem in frequentistic analysis using the sigmoid EMax model is the estimation of the Hill parameter, which may be problematic to estimate. The estimation problem for the Hill parameter and a model-based design approach to limit this problem will be examined in this paper.
KeywordsMaximum Likelihood Estimator Asymptotic Normality Fisher Information Matrix EMax Model Piecewise Linear Model
The author is grateful for very valuable discussions and input on this topic from Bjoern Bornkamp, Rainer Schwabe, Vladimir Dragalin and Valerii Fedorov.
- 2.Committee for Medicinal Products for Human Use (CHMP): Qualification opinion of MCP-Mod as an efficient statistical methodology for model-based design and analysis of Phase II dose finding studies under model uncertainty. European Medicines Agency, EMA/CHMP/SAWP/757052/2013 (2014)Google Scholar