Abstract

Ordinal response data are quite common in scientific experiments, and

finding the optimal design for them is a challenging task. Adjacent-category models are widely used to model ordinal response data. In this paper, we study the

D-optimal designs of adjacent-categories models with general link functions concerning both quantitative and qualitative factors. Some structure characteristics,

including the number of support points and a simple complete class of the locally

D-optimal design, are derived. Utilizing the obtained structure characteristics,

an efficient algorithm is proposed to search out corresponding D-optimal designs.

The integer-valued allocations for the corresponding D-optimal design are further

discussed for practical implementation. Numerical examples show the advantages

of the proposed design in both statistical efficiency and computational time.

Information

Preprint No.SS-2025-0075
Manuscript IDSS-2025-0075
Complete AuthorsHuiping Dang, Jun Yu, Fasheng Sun
Corresponding AuthorsFasheng Sun
Emailssunfs359@nenu.edu.cn

References

  1. Ai, M. Y., Z. Q. Ye and J. Yu (2023). Locally D-optimal designs for hierarchical response experiments. Statistica Sinica 33, 381–399.
  2. Agresti, A. (2019). An Introduction to Categorical Data Analysis. 3rd Edition. Wiley, New York.
  3. Bu, X. W., D. Majumdar and J. Yang (2020). D-optimal designs for multinomial logistic models. Annals of Statistics 48(2), 983–1000.
  4. Cheney, E. W. (1998). Introduction to Approximation Theory. 2nd Edition. American Mathematical Society.
  5. Chuang-Stein, C. and A. Agresti (1997). A review of tests for detecting a monotone doseresponse relationship with ordinal response data. Statistics in medicine 16, 2599–2618.
  6. Dette, H. and V. B. Melas (2011). A note on the de la garza phenomenon for locally optimal designs. Annals of Statistics 39(2), 1266–1281.
  7. Gevertz, J. L. and I. Kareva (2024). Minimally sufficient experimental design using identifiability analysis. Npj Systems Biology and Applications 10:2.
  8. Hao, S. and M. Yang (2020). Support point of locally optimal designs for multinomial logistic regression models. Journal of Statistical Planning and Inference 209, 144–159.
  9. Hedayat, A. S., N. J. Sloane and J. Stufken (1999). Orthogonal Arrays: Theory and Applications.
  10. Springer, New York.
  11. Huang, Y. F., K. Li, A. Mandal and J. Yang (2024). ForLion: a new algorithm for D-optimal designs under general parametric statistical models with mixed factors. Statistics and Computing 34:157.
  12. Huang, Y. F., S. T. Lin, and J. Yang (2025a). ForLion: ForLion Algorithm to Find D-Optimal
  13. Designs for Experiments. R package version 0.1.0.
  14. Huang, Y. F., L. P. Tong, and J. Yang (2025b). Constrained D-optimal design for paid research
  15. 0414−Preprint.pdf.
  16. Jennett, B. and M. Bond (1975). Assessment of outcome after severe brain damage. Lancet 305, 480–484.
  17. Karlin, S. and W. J. Studden (1966). TchebycheffSystems: With Applications in Analysis and Statistics. Pure and Applied Mathematics, Vol. XV. Wiley, New York.
  18. Kiefer, J. (1974). General equivalence theory for optimum designs (approximate theory). Annals of Statistics 2, 849–879.
  19. Lukemire, J., A. Mandal and W. K. Wong (2019). d-QPSO: a quantum-behaved particle swarm technique for finding D-optimal designs with discrete and continuous factors and a binary response. Technometrics 61(1), 77–87.
  20. Lukemire, J., A. Mandal and W. K. Wong (2022). Optimal experimental designs for ordinal models with mixed factors for industrial and healthcare applications. Journal of Quality Technology 54(2), 184–196.
  21. McCullagh, P. (1980). Regression models for ordinal data. Journal of the Royal Statistical Society, Series B 42(2), 109–142.
  22. McCullagh, P. and J. A. Nelder (1989). Generalized Linear Models. 2nd Edition. Chapman and Hall/CRC, Boca Raton.
  23. O’Connell, A. A. (2006). Logistic Regression Models for Ordinal Response Variables. 1st Edition. Sage.
  24. Ouyang, Y. Y., Y. M. Chen, G. H. Lan and E. Pasiliao (2015). An accelerated linearized alternating direction method of multipliers. SIAM Journal on Imaging Sciences 8, 644– 681.
  25. Perevozskaya, I., W. F. Rosenberger and L. M. Haines (2003). Optimal design for the proportional odds model. The Canadian Journal of Statistics 31, 225–235.
  26. Silvey, S. D. (1980). Optimal Design. Chapman & Hall/CRC.
  27. Smucker, B., M. Krzywinski and N. Altman (2018). Optimal experimental design. Nature Methods 15, 559–560.
  28. Stephen, J. W. (2015). Coordinate descent algorithms. Mathematical Programming 151, 3–34.
  29. Stufken, J. and M. Yang (2012). Optimal designs for generalized linear models. Design and Analysis of Experiments, Volume 3: Special Designs and Applications. Wiley.
  30. Yang, J., A. Mandal and D. Majumdar (2016). Optimal designs for 2k factorial experiments with binary response. Statistica Sinica 26, 385–411.
  31. Yang, J., L. P. Tong and A. Mandal (2017). D-optimal designs with ordered categorical data. Statistica Sinica 27, 1879–1902.
  32. Yang, M., B. Zhang and S. G. Huang (2011). Optimal designs for generalized linear models with multiple design variables. Statistica Sinica 21, 1415–1430.
  33. Zocchi, S. S. and A. C. Atkinson (1999). Optimum experimental designs for multinomial logistic models. Biometrics 55, 437–444. KLAS and School of Mathematics and Statistics, Northeast Normal University

Acknowledgments

Authors are grateful to the editor, associate editor, and the reviewers for

valuable comments and suggestions. Huiping Dang and Jun Yu contributed

Dang and Fasheng Sun’s research is supported by National Natural Science

Foundation of China (No. 12371259) and the Fundamental Research Funds

for the Central Universities (No. 2412023YQ003). Jun Yu’s research is supported by National Natural Science Foundation of China (No. 12471244).

Supplementary Materials

Supplement to “D-Optimal Designs for Ordinal Response Experiments”.

This supplementary material includes all technical proofs, an extension of

the AC po model and additional simulation studies.

DODORE-code. It contains codes for the simulation studies.

Supplementary materials are available for download.