The performance of the design matrices obtained from boundary and interior support points of a response surface is compared within a specified experimental area based on first order response surface model. This is achieved by determining the values of the desired optimal design criteria with respect to their corresponding information matrices. These optimality criteria are: D- optimality, G- optimality, A –optimality and E- optimality. Four designs arbitrarily chosen from both boundary and interior support points were used for the study. The designs obtained from boundary support points were found to be better than the designs obtained from interior support points with respect to the aforementioned design optimality criteria considered in this work.
This work by European American Journals is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License