A Validation of an Accounting Upper Error Limit Bound.
Abstract
The purpose of this thesis was to examine the validity of an accounting upper error limit bound. The bound examined was the DUS-cell method suggested by Leslie, Teitlebaum, and Anderson which was supposed to reduce bound conservatism and produce actual confidence levels closer to the nominal confidence levels. The analysis was accomplished by examining the robustness, the relative tightness, and the effects of error rate, error clustering, mean taint, and error amount intensity on the coverage and relative tightness of the DUS-cell bound. The results of this research indicate that the DUS-cell bound is not robust and is tighter than the Stringer bound. The results also demonstrated that error rate has the greatest effect on coverage and relative tightness, error clustering has some effect, and mean taint has little effect. The results also indicate that error amount intensity, a population characteristic, affects coverage and relative tightness of the DUS-cell bound significantly.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1985
- Accession Number
- ADA161205
Entities
People
- Michael W. Helton
Organizations
- Air Force Institute of Technology