Accession Number : AD0709052

Title :   ESTIMATION AND CONFIDENCE INTERVALS FOR QUANTAL RESPONSE OR SENSITIVITY DATA,

Corporate Author : SOUTHERN METHODIST UNIV DALLAS TEX DEPT OF STATISTICS

Personal Author(s) : Seibert,Guy Burton , Jr

Report Date : 25 FEB 1970

Pagination or Media Count : 69

Abstract : If different levels of stimuli are applied and then counts are made of the number reacting and not-reacting to the stimuli, a sensitivity analysis is usually the best analysis to be applied. Sensitivity analysis may be used to determine the probability of destroying a target with different size bombs, without the necessity of actually having data on all the different sizes. It may also be used in determining the effect of a drug on a patient or the effect of polution on a body of water. This paper is concerned with a generalized model for sensitivity analysis based on the Burr distribution. This generalization gives a model that will fit quite well data that comes from probit or logit models, but also covers many additional situations that the probit or logit models would not fit. A FORTRAN IV program is included for doing the analysis on a computer. Confidence intervals for any quantile are also given. (Author)

Descriptors :   (*BIOASSAY, EXPERIMENTAL DATA), (*STATISTICAL ANALYSIS, DECISION THEORY), CONFIDENCE LIMITS, STATISTICAL DISTRIBUTIONS, COMPUTER PROGRAMS, SENSITIVITY, THESES

Subject Categories : Biochemistry
      Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE