Abstract | ||
---|---|---|
The asymptotic confidence (delta) intervals and intervals based upon the use of Fieller’s theorem for effective doses were compared with two new approaches by Huang et al. (J. Statist. Comput. Simulation 72 (2002) 565) in a simulation study for a logistic dose–response curve. It was shown that these two novel approaches perform well to achieve a compromise between the conservative tendency of the Fieller interval and the liberalism often exhibited by the delta interval. Here attention is focused on the analytic comparison of the Fieller and delta intervals with one of two new methods, termed λ-interval, which includes the Fieller and delta intervals as special cases. The properties of this family of interval estimation methods are investigated, an analytical comparison of the various methods is provided and Sitter and Wu’s (J. Amer. Statist. Assoc. 88 (1993) 1021) analysis for explaining the difference in performance of the delta and Fieller intervals is extended to include the λ-interval based on simulation studies. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1016/j.csda.2004.05.004 | Computational Statistics & Data Analysis |
Keywords | Field | DocType |
Binary data,Effective doses,Interval estimators,Logistic model | Econometrics,Interval estimation,Statistical simulation,Binary data,Statistics,Confidence interval,Logistic regression,Mathematics,Estimator | Journal |
Volume | Issue | ISSN |
49 | 1 | 0167-9473 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yangxin Huang | 1 | 12 | 4.12 |