STA 6208 – Spring 2004 – Exam 1Print Name:UFID:All questions are based on the following two regression models, where SIMPLE REGRESSIONrefers to the case wherep=1, and X is of full column rank (no linear dependencies among thepredictor variables)Model 1:Yi=β0+β1Xi1+···+βpXip+εii,...,n εi∼NID(0,σ2)Model 2: Y=Xβ+εX≡n×p±β≡p±×1ε∼N(02I)Cochran’s TheoremSupposeYis distributed as follows with nonsingular matrixV:Y∼N(μ,Vσ2)r(V) =nthen:1.Y±(1σ2A)Yis distributed noncentralχ2with:(a) Degrees of freedom =r(A)(b) Noncentrality parameter = Ω =12σ2μ±AμifAVis idempotent2.Y±AYandY±BYare independent ifAVB=03.Y±and linear functionBYare independent ifBVA=01) Based onModel 1,derivethe normal equations for the simple linear regression model.2) Show that∑ni=1(ˆYi-Y) = 0. You may do this based on eitherModel 1orModel 2.3) A simple linear regression is Ft, relating Frst weekend revenues (Y) to advertising expenditures (X) forn= 5 randomly selected horror Flms:±ilmiSalesAd ExpScarier Movie125.08.0I Know What You Did Last Winter215.06.0Rural Legend312.04.0Shout430.010.0
This is the end of the preview. Sign up
access the rest of the document.
This note was uploaded on 07/28/2011 for the course STA 6208 taught by Professor Park during the Fall '08 term at University of Florida.