Interval estimation in risk analysis with nonquantal data
LE3 .A278 2017
Masters of Science
Mathematics & Statistics
In the literature of low-dose risk assessment, Piegorsch et al. (2005b) proposed five approaches to construct simultaneous confidence bounds with nonquantal data and they recommended Akahira’s Cornish-Fisher expansion method. In this thesis, a generalized confidence interval method proposed by Weerahandi (1995) is used to construct simultaneous confidence bounds for low-dose risk assessment when sample sizes arelarge. We apply small sample asymptotic methods (signed log likelihood ratio and modified signed log likelihood ratio) to obtain interval inference for risk assessment. Simulation studies are conducted to compare their performances based on the coverage probability. The application of the proposed methods is demonstrated by a real data example
The author grants permission to the University Librarian at Acadia University to reproduce, loan or distribute copies of my thesis in microform, paper or electronic formats on a non-profit basis. The author retains the copyright of the thesis.