Speaker: Timothy E. O’Brien, Loyola University Chicago

Title: Practical Design Strategies for Synergy-Interaction Modelling

Abstract: Researchers often find that generalized nonlinear regression models are applicable for bioassay assessment including relative potency and synergy modelling; in line with other biological, physical and chemical processes these nonlinear models perform better than linear models since they tend to fit the data well and the associated model parameter(s) are typically scientifically meaningful.  Generalized nonlinear model parameter estimation presents statistical modelling challenges including convergence and curvature (differential geometry) issues.  Researchers are also often in a position of requiring optimal or near-optimal (so-called “robust”) designs for the given chosen nonlinear model.  A common shortcoming of most optimal designs for nonlinear models used in practical settings, however, is that these designs typically focus only on (first-order) parameter variance or predicted variance, and thus ignore the inherent nonlinear of the assumed model function.  Another shortcoming of optimal designs is that they often have only p support points, where p is the number of model parameters.
 
This talk examines modelling and estimation methods connected with generalized nonlinear models useful for synergy modelling and provides concrete suggestions related to robust design strategies for these models.  Numerous examples will be provided and software methods and algorithms will be discussed.