This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: STAT 426 HW7. 1. (5.24) logit( π ) = α + β 1 A + β 2 S + β 3 R + β 4 R * S, where A =average number of alcoholic drink drinks consumed per day. S = braceleftBigg 1 , at least one pack per day , o.w. R = braceleftBigg 1 , black , white Y = braceleftBigg 1 , yes , no a. Prediction equation: when R = 1 , logit(ˆ π ) = (ˆ α + ˆ β 3 ) + ˆ β 1 A + ( ˆ β 2 + ˆ β 4 ) S = 6 . 7 + 0 . 1 A + 1 . 4 S when R = 0 , logit(ˆ π ) = ˆ α + ˆ β 1 A + ˆ β 2 S = 7 + 0 . 1 A + 1 . 2 S YS conditional odds ratio: When R = 1 : log(odds( s = 1)) = (ˆ α + ˆ β 2 + ˆ β 3 + ˆ β 3 + ˆ β 4 ) + ˆ β 1 A ——(1) log(odds( s = 0)) = (ˆ α + ˆ β 3 ) + ˆ β 1 A ——(2) (1)(2): log(OR(S))= ˆ β 2 + ˆ β 4 = 1 . 4 ⇒ OR(S) = exp(1 . 4) = 4 . 06 When R = 0 : log(odds( s = 1)) = (ˆ α + ˆ β 2 ) + ˆ β 1 A ——(3) log(odds( s = 0)) = ˆ α + ˆ β 1 A ——(4) (3)(4): log(OR(S))= ˆ β 2 = 1 . 2 ⇒ OR(S) = exp(1 . 2) = 3 . 32 Prediction equation: when S = 1 , logit(ˆ...
View
Full
Document
This note was uploaded on 04/29/2010 for the course STAT stat 426 taught by Professor Xe during the Spring '10 term at University of Illinois at Urbana–Champaign.
 Spring '10
 Xe

Click to edit the document details