• P-ISSN 0974-6846 E-ISSN 0974-5645

Indian Journal of Science and Technology


Indian Journal of Science and Technology

Year: 2023, Volume: 16, Issue: 39, Pages: 3407-3413

Original Article

A – Optimal Design for Poisson Regression Model Using Square Root Link

Received Date:07 August 2023, Accepted Date:13 September 2023, Published Date:25 October 2023


Objective: The main objective of this study is to identify the optimal operating conditions by determining the point of least variation on response variables, crucial for various aspects of product quality such as yield or strength. Method: Newton-Raphson method is used and the solution to these formulas are obtained numerically using R-software. The parameters in the context of A-optimal designs is a necessary step i.e., θ= (α, β, μ). Findings: Our findings show that A-optimal designs outperform conventional design criteria in terms of lowering parameter estimate variance. Moreover, we illustrate how the A-optimal designs can lead to a more efficient allocation of resources statistical models, optimizing the use of time and resources while ensuring robust statistical conclusions. Novelty and Applications : In the context of generalized linear models, most of the recent studies were on logistic regression models and many of them focused on optimal experimental designs with concentration on D optimality. In this research, Poisson regression models were considered for A-optimization with square root link. This approach proves to be particularly valuable as it accounts for the model's dependence on unknown parameters, making it a useful technique for practical applications. Suggestions: This methodology is quite general and may be applied to find A-optimal designs for other models, like the Compound Poisson, Gamma and inverse Gaussian model, with log link or other types of optimal designs.

Keywords: A­Optimality, Poisson Regression Model, Square Root Link, Experimental Design, Parameter Estimation


  1. Jung Y, Lee I. Optimal design of experiments for optimization-based model calibration using Fisher information matrix. Reliability Engineering & System Safety. 2021;216:107968. Available from: https://doi.org/10.1016/j.ress.2021.107968
  2. Belmiro PM, Duarte, Atkinson AC, José FO, Granjo, Nuno MC, et al. Optimal Design of Experiments for Implicit Models. Journal of the American Statistical Association. 2021;117(539):1424–1437. Available from: https://doi.org/10.1080/01621459.2020.1862670
  3. Liu L, Li J, Deng Y, Yang Z, Huang K, Zhao S. Optimal design of multi-layer structure composite containing inorganic hydrated salt phase change materials and cement: Lab-scale tests for buildings. Construction and Building Materials. 2021;275:122125. Available from: https://doi.org/10.1016/j.conbuildmat.2020.122125
  4. Borrotti M, Sambo F, Mylona K. Multi-objective optimisation of split-plot designs. Econometrics and Statistics. 2023;28:163–172. Available from: https://doi.org/10.1016/j.ecosta.2022.04.001
  5. García-Ródenas R, García-García JC, López-Fidalgo J, Martín-Baos JÁ, Wong WK. A comparison of general-purpose optimization algorithms for finding optimal approximate experimental designs. Computational Statistics & Data Analysis. 2020;144:106844. Available from: https://doi.org/10.1016/j.csda.2019.106844
  6. Wong WK, Zhou J. Using CVX to Construct Optimal Designs for Biomedical Studies with Multiple Objectives. Journal of Computational and Graphical Statistics. 2022;32(2):744–753. Available from: https://doi.org/10.1080/10618600.2022.2104858
  7. Goos P, Syafitri U, Sartono B, Vazquez AR. A nonlinear multidimensional knapsack problem in the optimal design of mixture experiments. European Journal of Operational Research. 2020;281(1):201–221. Available from: https://doi.org/10.1016/j.ejor.2019.08.020
  8. Idais O. Locally optimal designs for multivariate generalized linear models. Journal of Multivariate Analysis. 2020;180:104663. Available from: https://doi.org/10.1016/j.jmva.2020.104663
  9. Grahoff U, Holling H, Schwabe R. D‐optimal design for the Rasch counts model with multiple binary predictors. British Journal of Mathematical and Statistical Psychology. 2020;73(3):541–555. Available from: https://doi.org/10.1111/bmsp.12204
  10. Schmidt M, Schwabe R. Optimal designs for Poisson count data with Gamma block effects. Journal of Statistical Planning and Inference. 2020;204:128–140. Available from: https://doi.org/10.1016/j.jspi.2019.05.002
  11. Chen PY, Chen CC, Chen YS, Kee W, Wong. Numerical Methods for Finding A-optimal Designs Analytically. Econometrics and Statistics. 2022. Available from: https://doi.org/10.1016/j.ecosta.2022.09.005
  12. Jones B, Allen-Moyer K, Goos P. A-optimal versus D-optimal design of screening experiments. Journal of Quality Technology. 2021;53(4):369–382. Available from: https://doi.org/10.1080/00224065.2020.1757391
  13. Niaparast M, Mehrmansour S, Schwabe R. V-optimality of designs in random effects Poisson regression models. Metrika. 2023;86(8):879–897. Available from: https://doi.org/10.1007/s00184-023-00896-3
  14. Olamide EI, Fasoranbaku OA, Adebola FB. E-optimal Experimental Designs for Poisson Regression Models in Two and Three Variables. Tanzania Journal of Science. 2021;47(3):999–1006. Available from: https://doi.org/10.4314/tjs.v47i3.11
  15. Jiang H, Yue R. D-optimal population designs in linear mixed effects models for multiple longitudinal data. Statistical Theory and Related Fields. 2021;5(2):88–94. Available from: https://doi.org/10.1080/24754269.2021.1884444
  16. Goudarzi M, Khazaei S, Jafari H. D-optimal designs for linear mixed model with random effects of Dirichlet process. Communications in Statistics - Simulation and Computation. 2021;p. 1–10. Available from: https://doi.org/10.1080/03610918.2021.1987467
  17. Masoudi E, Holling H, Duarte BPM, Wong WK. A Metaheuristic Adaptive Cubature Based Algorithm to Find Bayesian Optimal Designs for Nonlinear Models. Journal of Computational and Graphical Statistics. 2019;28(4):861–876. Available from: https://doi.org/10.1080/10618600.2019.1601097
  18. Feng Y, Liu Q, Yao Q, Zhao G. Model Averaging for Nonlinear Regression Models. Journal of Business & Economic Statistics. 2022;40(2):785–798. Available from: https://doi.org/10.1080/07350015.2020.1870477
  19. Li K, Yang J. Score-matching representative approach for big data analysis with generalized linear models. Electronic Journal of Statistics. 2022;16(1):16. Available from: https://doi.org/10.1214/21-ejs1965
  20. Brown VA. An Introduction to Linear Mixed-Effects Modeling in R. Advances in Methods and Practices in Psychological Science. 2021;4(1). Available from: https://doi.org/10.1177/2515245920960351


© 2023 Swetha & David. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)


Subscribe now for latest articles and news.