Optimal designs for generalized non‐linear models with application to second‐harmonic generation experiments

The design of experiments for generalized non-linear models is investigated and applied to an optical process for characterizing interfaces which is widely used in the physical and natural sciences. Design strategies for overcoming the dependence of a D-optimal design on the values of the model parameters are explored, including the use of Bayesian designs. Designs for the accurate estimation of model parameters are presented and compared, as are designs for the estimation of a set of ratios of parameters which is of particular importance in the motivating example. The effectiveness of various design methods is studied, and the benefits of well-designed experiments are demonstrated.

[1]  V. Melas.  On the functional approach to optimal designs for nonlinear models , 2005 .

[2]  P. McCullagh,et al.  Generalized Linear Models , 1972, Predictive Analytics.

[3]  Michel Wedel,et al.  Heterogeneous Conjoint Choice Designs , 2005 .

[4]  John E. Sipe,et al.  Phenomenological treatment of surface second-harmonic generation , 1988 .

[5]  John E. Dennis,et al.  Numerical methods for unconstrained optimization and nonlinear equations , 1983, Prentice Hall series in computational mathematics.

[6]  T. Pham-Gia,et al.  Distributions of the ratios of independent beta variables and applications , 2000 .

[7]  A. Welsh,et al.  Statistical analysis of second harmonic generation experiments: a phenomenological model , 2005 .

[8]  Susan M. Lewis,et al.  D-OPTIMAL DESIGNS FOR POISSON REGRESSION MODELS , 2009 .

[9]  R. W. Wedderburn.  Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method , 1974 .

[10]  David C. Woods,et al.  Designs for Generalized Linear Models With Several Variables and Model Uncertainty , 2006, Technometrics.

[11]  Anthony C. Atkinson,et al.  Optimum Experimental Designs, with SAS , 2007 .

[12]  K. Chaloner,et al.  Optimal Bayesian design applied to logistic regression experiments , 1989 .

[13]  J. Frey,et al.  The temperature dependence of surface second-harmonic generation from the air-water interface , 2001 .

[14]  Leslie M Loew,et al.  Second-harmonic imaging microscopy for visualizing biomolecular arrays in cells, tissues and organisms , 2003, Nature Biotechnology.

[15]  M. Wedel,et al.  Designing Conjoint Choice Experiments Using Managers' Prior Beliefs , 2001 .

[16]  Bo-Cheng Wei,et al.  Exponential Family Nonlinear Models , 1998 .

[17]  F. Pukelsheim,et al.  Efficient rounding of approximate designs , 1992 .

[18]  R. Serfling.  Approximation Theorems of Mathematical Statistics , 1980 .

[19]  David Firth,et al.  GNM: a package for generalized nonlinear models , 2006 .

[20]  David Firth,et al.  Generalized nonlinear models in R: An overview of the gnm package , 2007 .

[21]  A. J. D. Dekker,et al.  Resolution of coherent and incoherent imaging systems reconsidered - Classical criteria and a statistical alternative , 2006 .

[22]  D. Van Dyck,et al.  Optimal experimental design of STEM measurement of atom column positions. , 2002 .

[23]  J. Salafsky.  Second-harmonic generation for studying structural motion of biological molecules in real time and space. , 2007, Physical chemistry chemical physics : PCCP.

[24]  Y. Shen,et al.  Surface properties probed by second-harmonic and sum-frequency generation , 1989, Nature.

[25]  David M. Steinberg,et al.  Fast Computation of Designs Robust to Parameter Uncertainty for Nonlinear Settings , 2009, Technometrics.

[26]  Hovav A. Dror,et al.  Robust Experimental Design for Multivariate Generalized Linear Models , 2006, Technometrics.

[27]  David Firth,et al.  Bias reduction in exponential family nonlinear models , 2009 .