This preview shows pages 1–3. 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: 620-202 Tutorial/Computing Labo-ratory 5.We consider the simple linear modelY=+(xi-x) +wheren(0,2).1. (6.11-6) Show thatnXi=1[Yi--(xi-x)]2=n(-)2+ (-)2nXi=1(xi-x)2+nXi=1[Yi--(xi-x)]22. (6.11-7)Show that the endpoints for a 100(1-)% confidence interval foraret/2(n-2)sb2n-23. (6.11-8)Show that a 100(1-)% confidence interval for2is"nb22/2(n-2),nb221-/2(n-2)#4. (6.12-11) Explain why the model(x) =1exp(2x) is not a linear model.5. (6.12-12) To fit the quadratic curvey=1+2x+3x2to a set of points weminimizeh(1,2,3) =nXi=1(yi-1-2xi-3x2i)2By setting the three first partial derivatives ofhwith respect to1,2and3tozero show that1,2and3satisfy the normal equationsXyi=1n+2Xxi+3Xx2iXxiyi=1Xxi+2Xx2i+3Xx3iXx2iyi=1Xx2i+2Xx3i+3Xx4i16. (6.11- 5) (R problem). A student was interested in how horse power and weightof a car affected the time it takes a car to go from 0 tp 60 mph. The followingtable gives for 14 cars the horsepower, time in seconds to go from 0-60 mph and...
View Full Document
This note was uploaded on 10/05/2009 for the course STATS 620-202 taught by Professor R during the One '09 term at University of Melbourne.
- One '09