Quantile dispersion graphs for evaluating and comparing designs for logistic regression models

Designs for fitting a generalized linear model depend on the unknown parameters of the model. The use of any design optimality criterion would therefore require some prior knowledge of the parameters. In this article, a graphical technique is proposed for comparing and evaluating designs for a logistic regression model. Quantiles of the scaled mean-squared error of prediction are obtained on concentric surfaces inside a region of interest, R. For a given design, these quantiles depent on the model's parameters. Plots of the maxima and minima of the quantiles, over a subset of the parameter space, produce the so-called quantile dispersion graphs (QDGs). The plots provide a comprehensive assessment of the overall prediction capability of the design within the region R. They also depict the dependence of the design on the model's parameters. The QDGs can therefore be conveniently used to compare several candidate designs. Two examples are presented to illustrate the proposed methodology.

[1]  Gilberto A. Paula,et al.  Bias correction for exponential family nonlinear modles , 1992 .

[2]  W. Piegorsch.  AN INTRODUCTION TO BINARY RESPONSE REGRESSION AND ASSOCIATED TREND ANALYSES , 1998 .

[3]  David W. Scott.  The New S Language , 1990 .

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

[5]  D. Cox,et al.  A General Definition of Residuals , 1968 .

[6]  P. McCullagh,et al.  Bias Correction in Generalized Linear Models , 1991 .

[7]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[8]  K. Chaloner,et al.  Bayesian design for accelerated life testing , 1992 .

[9]  Elizabeth A. Peck,et al.  Introduction to Linear Regression Analysis , 2001 .

[10]  Vernon M. Chinchilli,et al.  An Asymptotic Confidence Region for the ED 100p From the Logistic Response Surface for a Combination of Agents , 1986 .

[11]  R. H. Myers,et al.  A TUTORIAL ON GENERALIZED LINEAR MODELS , 1997 .

[12]  K. Chaloner,et al.  Bayesian Experimental Design: A Review , 1995 .

[13]  André I. Khuri,et al.  A graphical approach for evaluating and comparing designs for nonlinear models , 1998 .

[14]  Linda J. Young,et al.  Estimation of Extreme Quantiles Based on Sensitivity Tests: A Comparative Study , 1994 .

[15]  Linda M. Haines,et al.  Bayesian D-optimal designs for the exponential growth model , 1995 .

[16]  Thomas P. Hettmansperger,et al.  Robust procedures for drug combination problems with quantal responses , 1992 .

[17]  Chih-Ling Tsai,et al.  Bias in nonlinear regression , 1986 .

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

[19]  N. G. Cadigan,et al.  Bias approximation for maximum likelihood estimates , 1994 .

[20]  P. McCullagh.  Tensor Methods in Statistics , 1987 .