Precision of full polynomial response surface designs on models with missing coefficients

  • Authors

    • Mary Iwundu Department of Mathematics and Statistics, University of Port Harcourt, Nigeria
    • ENYINNA Jaja Department of Mathematics and Statistics, Rivers State University, Nigeria
    2017-05-03
    https://doi.org/10.14419/ijasp.v5i1.7491
  • Condition Number, Design Efficiency, Full Models, Reduced Models and Response Surface Designs.

  • The precision of using full polynomial response surface designs on models with missing coefficients (reduced models) is studied using efficiency measures. The loss in D- and G-efficiency of constructed first-order exact designs is minimized for the model with missing interaction coefficient. However, higher losses in D- and G-efficiency are recorded when constructed second-order exact designs are used on the model with missing interaction coefficient with few exceptions showing preferences for using the designs on the reduced model. Lower condition numbers are observed for the designs under the first-order reduced models thus indicating that the N-point exact designs are closer to being orthogonal for the reduced model than for the full model. Perfect orthoganality is achieved at design sizes 4 and 8. In fact, N-point exact designs of multiples of N=4 show perfect orthoganality when defined either for the full or reduced first-order models. In comparison to a design with perfect orthoganality, the second-order designs are far from being orthogonal.

  • References

    1. [1] J. E. Bon, Generating exact D-optimal designs for polynomial model, SpringSim, 2 (2007) 121-126.

      [2] R. B. Crosier, Response Surface design comparisons: Technical Report. U. S. Army Edgewood Research, Development and Engineering Center, SCBRD-RTM, Bldg. E3160 Aberdeen Proving ground, Maryland, MD 21010-5423 (1993).

      [3] J. Fan, W. Zhang, Statistical methods with varying coefficient models, Statistics and its Interface, 1(2008) 179-195. https://doi.org/10.4310/SII.2008.v1.n1.a15.

      [4] V. V. Fedorov, Theory of Optimal Experiments, Academic Press, New York, 1972.

      [5] M. L. Huang, F. Chang, W. K. Wong, D-optimal designs for polynomial regression without an intercept. Statistica Sinica, 5 (1995) 441-458.

      [6] M. P. Iwundu, E. B. Albert-Udochukwuka, On the Behaviour of D-Optimal Exact Designs Under Changing Regression Polynomials, International Journal of Statistics and Probability, 3(4)(2014) 67-85. doi:10.5539/ijsp.v3n4p67 URL: http://dx.doi.org/10.5539/ijsp.v3n4p67.

      [7] M. P. Iwundu, P. E. Chigbu, A hill-climbing Combinatorial Algorithm for constructing N-point D-optimal exact designs, Journal of Statistics Applications and Probability, 1(2) (2012) 133-146. https://doi.org/10.12785/jsap/010205.

      [8] A. I. Khuri, S. Mukhopadhyay, Response Surface Methodology: Advanced Review, WIREs Computational Statistics, 2 (2010)128-149. https://doi.org/10.1002/wics.73.

      [9] R. H. Myers, D. C. Montgomery, C. M. Anderson-Cook, Response Surface Methodology: Process and Product Optimization using designed experiments, 3rd Edition, John Wiley & Sons, Inc. New Jersey, 2009.

      [10] I. B. Onukogu, Foundation of optimal exploration of response surfaces, Ephrata Press, Nsukka, Nigeria, 1997.

      [11] K. Toh, W. Yah, X. Jiang, A reduced multivariate polynomial model for Multivariate Biometrics and Classifiers fusion, IEEE Transactions on Circuits and Systems for video technology, 14(2) (2004) 224-233. https://doi.org/10.1109/TCSVT.2003.821974.

      [12] H. P. Wynn, The sequential generation of D-optimal experimental designs, The Annals of Mathematical Statistics, 41(5) (1970) 1655-1664. https://doi.org/10.1214/aoms/1177696809.

      [13] Y. Yuan, H. Zhu, M. Styner, J. H. Gilmore, J. S. Marron, Varying coefficient model for modeling diffusion tensors along white matter tracts, Annals of Applied Statistics, 7(1) (2013) 102-125. https://doi.org/10.1214/12-AOAS574.

      [14] H. Zhu, R. Li, L. Kong, Multivariate varying coefficient model for functional Responses, Annals of Statistics, 40(5) (2012)2634-2666. https://doi.org/10.1214/12-AOS1045.

  • Downloads