On Asymptotically Distribution-Free Confidence Bounds for P (X1≥X2) Based on Samples not Necessarily Independent
✍ Scribed by Dr. R. Hilgers
- Publisher
- John Wiley and Sons
- Year
- 1981
- Tongue
- English
- Weight
- 278 KB
- Volume
- 23
- Category
- Article
- ISSN
- 0323-3847
No coin nor oath required. For personal study only.
✦ Synopsis
If Xl, X2 denote the random variables of measurement under two treatments then the probability P (XI h X2) is a quantity of great practical interest. eapecially if we consider both to be measured for the name unity. In, this caae the random variables cannot be assumed to be indepeudent any longer. The following paper describes a procodure to compute approximate confidence bounds for P ( X , z X 2 ) where correlatiom between xl, x2 are a d m i t t e d 88 well as between replications of the Xf. There is some relation to the FRIEDMAY-8tatiStk with or without repeated measurements and as a special cmae to the sign-test. Application may be extended to ordinal d8ta.
📜 SIMILAR VOLUMES