Robustness of Augmented Third-order Response Surfaces Designs to Missing Observation

Document Type : Original Article

Authors
Yankam Mbouamba Brenda
Abimibola Victoria Oladugba
Department of Statistics, Faculty of Physical Sciences, University of Nigeria, Nsukka, Nigeria.
10.22034/jirss.2024.2015205.1044
Abstract
Missing observations are practical problems that occur frequently even in a well-planned experiment and can significantly impact the statistical accuracy of the experiment. This work introduces a new class of third-order designs called augmented orthogonal uniform composite minimax loss (AOUCM) designs, which are more robust to a single missing design point as a variation of the existing third-order augmented orthogonal uniform composite designs (AOUCDs). The AOUCM designs are constructed using the minimax loss criterion. The constructed AOUCM designs are evaluated and compared with AOUCDs based on the relative D- and G-efficiency criteria, generalized scaled deviation, and the fraction of design space plot. The AOUCM designs are shown to be robust and more efficient in estimating the parameters of the third-order model. Moreover, although the AOUCDs and AOUCM designs are stable and uniformly distributed throughout the design space, the AOUCM designs have the least scaled prediction variance.
Keywords
Third-order Models
Augmented Orthogonal Uniform Composite Designs
Minimax Loss Criterion
Missing Observation
Relative Efficiency
Fraction of Design Space Plot
Subjects
Ahmad, T., Akhtar, M., and Gilmour, S. G. (2012). Multilevel augmented pairs second-order response surface designs and their robustness to missing data.  Communications in Statistics - Theory and Methods, 41(3), 437–452.
Ahmad, T., and Akhtar, M. (2014). Efficient response surface designs for the second-order multivariate polynomial model robust to missing observation.  Journal of Statistical Theory and Practice, 9(2), 361–375.
Ahmad, T., and Gilmour, S. G. (2010). Robustness of subset response surface designs to missing observations.  Journal of Statistical Planning and Inference, 140(1), 92–103.
Akhtar, M., and Prescott, P. (1986). Response surface designs robust to missing observations.  Communications in Statistics - Simulation and Computation, 15(2), 345–363.
Alrweili, H., Georgiou, S., and Stylianou, S. (2019). Robustness of response surface designs to missing data.  Quality and Reliability Engineering International, 35(5), 1288–1296.
Arshad, H. M., Ahmad, T., and Akhtar, M. (2018). Some sequential third order response surface designs.  Communications in Statistics: Simulation and Computation, 49(7), 1872–1885.
Arshad, H. M., Akhtar, M., and Gilmour, S. G. (2012). Augmented Box-Behnken designs for fitting third-order response surfaces.  Communications in Statistics - Theory and Methods, 41(23), 4225–4239.
Chen, X. P., Guo, B., Liu, M. Q., and Wang, X. L. (2018). Robustness of orthogonal-array based composite designs to missing data.  Journal of Statistical Planning and Inference, 194, 15–24.
Cornell, J. A., and Montgomery, D. C. (1996). Interaction models as alternatives to low-order polynomials.  Journal of Quality Technology, 28(2), 163–176.
Draper, N. R. (1962). Third order rotatable designs in three factors: Analysis’.  Technometrics, 4(2), 219–234.
Fang, J., He, Z., He, S., and Wang, G. (2019). The robustness of response surface designs with errors in factor levels.  Communications in Statistics - Theory and Methods, 49(10), 2365–2383.
Hemavathi, M., Varghese, E., Shekhar, S., Athulya, C. K., Ebeneezar, S., Gills, R., and Jaggi, S. (2022). Robustness of sequential third-order response surface design to missing observations.  Journal of Taibah University for Science, 16(1), 270-279.
Marina, S. (2013). Dealing with missing data: Key assumptions and methods for applied analysis. School of Public Health Department, Boston University. Technical Report No. 4.
Oladugba, A. V., and Nwanonobi, O. C. (2021). Robustness of definitive screening composite designs to missing observations.  Communications in Statistics-Theory and Methods, 1-15.
Oladugba, A. V., and Yankam, B. M. (2021). Prediction variance properties of third-order response surface designs in the hypersphere.  Mathematics and Statistics, 10(1): 61-73.
Ossai, E. O., and Oladugba, A. V. (2018). The estimation of missing values in rectangular lattice designs.  Journal of Modern Applied Statistical Methods, 17(2), eP2670.
Oladugba, A. V., Ossai E. O., and Ugah T. E.(2018). On the Comparison of methods of estimating missing values in rectangular lattice designs.  Journal of Mathematics and Statistics, {\bf14(1), 201–208.
Rashid, F., Akbar, A., and Zafar, Z. (2019). Some new third order designs robust to one missing observation.  Communications in Statistics—Theory and Methods, 1–9.
Xu, H., Jaynes, J., and Ding, X.(2014). Combining two-level and three-level orthogonal arrays for factor screening and response surface exploration.  Statistica Sinica, {\bf24(1), 269–289.
Yankam, B. M., and Oladugba, A. V. (2021a). Robustness of orthogonal uniform composite designs against missing data.  Communications in Statistics - Theory and Methods, 1–16.
Yankam,B. M., and Oladugba, A. V.(2021b). Prediction variance capabilities of third-order response surface designs for cuboidal regions.  Mathematics and Statistics, {\bf9(5), 760-772.
Yankam,B.M., and Oladugba,A.V.(2022). Augmented orthogonal uniform composite designs for fitting third-order model. Journal of Statistical Theory and Practice,{\bf16(2),1–14.
Zahran, A., Anderson-Cook, C. M., and Myers, R. H. (2003). Fraction of design space to assess prediction capability of response surface designs.  Journal of Quality Technology, {\bf35(4), 377–386.
Zhang, X. R., Liu, M., and Zhou, Y. (2020). Orthogonal uniform composite designs.  Journal of Statistical Planning and Inference, {\bf206, 100–110.
Zhang, X. R., Qi, Z. F., Zhou, Y. D., and Yang, F. (2018). Orthogonal-array composite design for the third-order models.  Communications in Statistics - Theory and Methods, {\bf47(14), 3488–3507.
Journal of the Iranian Statistical Society
Volume 23, Issue 1
June 2024
Pages 33-49

  • Receive Date 06 November 2023
  • Revise Date 05 April 2024
  • Accept Date 06 July 2024