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: ISYE 6414  Spring 2009 Solution 1  The Linear Model: OLS Regression Problem 1. Maximun Likelihood estimators If Y i Normal ( + 1 x i , 2 ) , i = 1 , 2 ,...n , and Cov ( Y i ,Y j ) = 0, i 6 = j ; then the likelihood function is: L ( Y 1 ,...,Y n  , 1 ) = ( 2 2 ) n 2 exp n i =1 ( Y i  1 x i ) 2 2 2 ! Writing Y = ( Y 1 ,...,Y n ), the loglikelihood can be expressed as: l ( Y  , 1 ) = n 2 ln ( 2 2 ) n i =1 ( Y i  1 x i ) 2 2 2 (1) d l ( Y  , 1 ) d = 0 n X i =1 ( Y i  1 x i ) = 0 (2) d l ( Y  , 1 ) d 1 = 0 n X i =1 x i ( Y i  1 x i ) = 0 (3) Then, from equations (2) and (3) we get: n X i =1 Y i = nb + b 1 n X i =1 x i (4) n X i =1 Y i x i = b n X i =1 x i + b 1 n X i =1 x 2 i (5) And these are the normal equations for the OLS estimators. Problem 1.5 Pag.33 No, we know that the error term i is random. That implies that E ( Y i ) = + 1 x i + E ( i ) = + 1 x i and Y i = + 1 x i + i Problem 1.6 Pag.33 PLot... = 200: The expected value of the response Y when X = 0 is 200. 1 = 5: When X increases by one, the response Y is expected to increase by 5. Problem 1.7 Pag.33 a. In model (1.1) the exact probability cannot be determined unless some probability distribution is used to model the random term . b. In model (1.24), we know that i N ( , 2 = 25), then: P (195 Y 205  x = 5) = P ( 1 Z 1) Where Z N (0 , 1). Problem 1.8 Pag.33 Yes, the mean of the new observation Y * given X = 45 will still be E ( Y * ) = 9 . 5 + 2 . 1 45 = 104. However, the value of Y * is nor necessary 108 given that is a random vaiable....
View
Full
Document
This note was uploaded on 09/01/2011 for the course ISYE 6414 taught by Professor Staff during the Fall '08 term at Georgia Institute of Technology.
 Fall '08
 Staff

Click to edit the document details