ABSTRACT

Dose–response experiments and data analyses are often carried out according to an optimal design under a model assumption. A two-parameter logistic model is often used because of its nice mathematical properties and plausible stochastic response mechanisms. There is an extensive literature on its optimal designs and data analysis strategies. However, a model is at best a good approximation in a real-world application, and researchers must be aware of the risk of model mis-specification. In this paper, we investigate the effectiveness of the sequential EDdesign, the D-optimal design, and the up-and-down design under the three-parameter logistic regression model, and we develop a numerical method for the parameter estimation. Simulations show that the combination of the proposed model and the data analysis strategy performs well. When the logistic model is correct, this more complex model has hardly any efficiency loss. The three-parameter logistic model works better than the two-parameter logistic model in the presence of model mis-specification.

• Brown, C. C. (1982). On a goodness of fit test for the logistic model based on score statistics. Communications in Statistics - Theory and Methods, 11(10), 1087–1105. https://doi.org/10.1080/03610928208828295 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
• El-Saidi, M. A. (1993). A power transformation for generalized logistic response function with application to quantal bioassay. Biometrical Journal, 35(6), 715–726. https://doi.org/10.1002/(ISSN)1521-4036 doi: https://doi.org/10.1002/bimj.4710350609 [Crossref], [Web of Science ®], [Google Scholar]
• Fedorov, V. V. (1972). Theory of optimal experiments. Elsevier. [Google Scholar]
• Ford, I., Titterington, D. M., & Wu, C. F. J. (1985). Inference and sequential design. Biometrika, 72(3), 545–551. https://doi.org/10.1093/biomet/72.3.545 [Crossref], [Web of Science ®], [Google Scholar]
• Li, G., & Majumdar, D. (2008). D-optimal designs for logistic models with three and four parameters. Journal of Statistical Planning and Inference, 138(7), 1950–1959. https://doi.org/10.1016/j.jspi.2007.07.010 [Crossref], [Web of Science ®], [Google Scholar]
• Li, P., & Wiens, D. P. (2011). Robustness of design in dose-response studies. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 73(2), 215–238. https://doi.org/10.1111/rssb.2011.73.issue-2 doi: https://doi.org/10.1111/j.1467-9868.2010.00763.x [Crossref], [Google Scholar]
• O'Brien, T. E., Chooprateep, S., & Homkham, N. (2009). Efficient geometric and uniform design strategies for sigmoidal regression models. South African Statistical Journal, 43(1), 49–83. https://journals.co.za/content/sasj/43/1/EJC119263. [Google Scholar]
• Sitter, R. R., & Fainaru, I. (1997). Optimal designs for the logit and probit models for binary data. Canadian Journal of Statistics, 25(2), 175–190. https://doi.org/10.2307/3315730 [Crossref], [Web of Science ®], [Google Scholar]
• Sitter, R. R., & Forbes, B. (1997). Optimal two-stage designs for binary response experiments. Statistica Sinica, 7(4), 941–955. https://doi.org/10.2307/3315730. [Web of Science ®], [Google Scholar]
• Sitter, R. R., & Wu, C. (1993). Optimal designs for binary response experiments: Fieller, D, and A criteria. Scandinavian Journal of Statistics, 20(4), 329–341. [Web of Science ®], [Google Scholar]
• Wang, L., Liu, Y., Wu, W., & Pu, X. (2013). Sequential LND sensitivity test for binary response data. Journal of Applied Statistics, 40(11), 2372–2384. https://doi.org/10.1080/02664763.2013.817546 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
• Wu, C. (1978). Some algorithmic aspects of the theory of optimal designs. The Annals of Statistics, 6(6), 1286–1301. https://doi.org/10.1214/aos/1176344374 [Crossref], [Web of Science ®], [Google Scholar]
• Wu, C. F. J. (1985). Efficient sequential designs with binary data. Journal of the American Statistical Association, 80(392), 974–984. https://doi.org/10.1080/01621459.1985.10478213 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
• Wu, C. F. J., & Tian, Y. (2014). Three-phase optimal design of sensitivity experiments. Journal of Statistical Planning and Inference, 149, 1–15. https://doi.org/10.1016/j.jspi.2013.10.007 [Crossref], [Web of Science ®], [Google Scholar]
• Wynn, H. P. (1972). Results in the theory and construction of D-optimum experimental designs. Journal of the Royal Statistical Society, Series B, 34(2), 133–147. [Google Scholar]
• Yu, X., Chen, J., & Brant, R. (2016). Sequential design for binary dose-response experiments. Journal of Statistical Planning and Inference, 177, 64–73. https://doi.org/10.1016/j.jspi.2016.04.005 [Crossref], [Web of Science ®], [Google Scholar]

To cite this article: Xiaoli Yu, Shaoting Li & Jiahua Chen (2021) A three-parameter logistic regression model, Statistical Theory and Related Fields, 5:3, 265-274, DOI: 10.1080/24754269.2020.1796098

To link to this article: https://doi.org/10.1080/24754269.2020.1796098