Sensitivity of Design Optimality Criteria to Second-Order Response Surface Model Restrictions and Ce.docVIP

Sensitivity of Design Optimality Criteria to Second-Order Response Surface Model Restrictions and Ce 　　In this work, we investigate the role of several model characteristics and center point replications on the properties of A-, D-, G-, andIV- optimality for four?factor second-order response surface design. It was discovered thatA-, D-, and G- efficiencies tend to reduce as the center points are replicated while the effect on scaled average prediction variance tends to be very insignificant. Among the restricted models considered, the pure linear model (model1) turns out to be the best in terms of quality of estimation and model prediction. Angela U. Chukwu, Yisa Yakubu (2012). Sensitivity of Design Optimality Criteria to Second-order Response Surface Model Restrictions and Center Point Replications. Studies in Mathematical Sciences, 4(1), 22-29. Available from: URL: / index.php/sms/article/view/j.sms.1923845220120401.1522 DOI: /10.3968/j.sms.1923845220120401.1522 　　 INTRODUCTION 　　Experiments are performedin virtually allfields of inquiry usually to discover something about a particular process or system. Designed experiments allow the analyst to control the factors thought to be important in characterizing or explaining the response variable(s) of the experiment. Response surface methodology(RSM) is an area of experimental design which consists of a group of mathematical and statistical techniques used in the developmentof an adequatefunctionalrelationship between a response of interest, y, and a number of associated input variables (Montgomery D. C., 2001).Usually, the form of the relationship is unknown but can be approximated,within the experimental region, by a low degree polynomial such as the second?order response surface model 　　where y represents the measured response,?βs are parameter coefficients and e is anterm(independentlyand normally distributed with mean zero and common varianceσ2e) that accounts for random error and bias(Khuri, A. I. and Co