|
Activity Number:
|
333
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #309248 |
|
Title:
|
Fiducial Generalized Confidence Interval for Median Lethal Dose (LD50)
|
|
Author(s):
|
Lidong E*+ and Jan Hannig and Hariharan Iyer
|
|
Companies:
|
Colorado State University and Colorado State University and Colorado State University
|
|
Address:
|
120 Marble Street 302, Broomfield, CO, 80020,
|
|
Keywords:
|
LD50 ; logistic model ; Fiducial Generalized Confidence Interval ; Gibbs sampling
|
|
Abstract:
|
In this paper, we focus on the construction of confidence intervals for the median lethal dose LD50, which is a common measure of the acute toxicity of a compound in a species. We propose a new method for constructing confidence intervals of LD50 for a logistic-response curve. This method is based on an extension of R. A. Fisher's fiducial argument developed by Hannig (2006). The Gibbs sampling approach is used to empirically estimate the percentiles of the fiducial distribution for LD50. The resulting intervals are compared with three other competing confidence interval procedures - Delta method, Fieller's method and Likelihood Ratio method. Our simulation results show that fiducial intervals have satisfactory performance and are more stable than other confidence intervals in terms of coverage probability. The fiducial distributions also appear to give unbiased point estimates of LD50.
|
