This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Testing a Normal Mean Earlier found sample size to achieve specified Type I error and Type II error at particular value of alternative parameter. Here n given. Cant pick it. Stuck with whatever emerges. Test at the level of significance H : = H 1 : < 1 Recall CR given by { x < c } , c determined by c 30 / n = z that is c =  z n Decision: Reject H if X < c. Meaning: Sample average must be a certain number of stan dard error units away from the null value (similar to being guilty beyond a reasonable doubt in a criminal trial.) 2 Equivalently compute standard ized z score or signal to noise ratio z = x / n Reject H if z < z . Note . Can replace H : = with composite null hypothesis H : because from OC curves of previous lec ture, largest Type I error occurs at boundary between H , H 1 . 3 Example . Data 40 26 39 14 42 18 25 43 46 27 19 47 19 26 35 34 15 44 40 38 31 46 52 25 35 35 33 29 34 41 49 28 52...
View
Full
Document
This note was uploaded on 01/03/2012 for the course EE 1244 taught by Professor Drera during the Fall '10 term at Conestoga.
 Fall '10
 drera

Click to edit the document details