Quiz_2_Solutions - MSE ( ) = E ( - ) 2 = V ar ( ) + bias (...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
IEOR 165: Engineering Statistics, Quality Control and Forecasting, Spring 2007 Quiz 2 Solution Question 1 (a) The expected value of the first moment for a single uniform random variable E [ X ] = Z b -∞ xe x - b dx = Z 0 -∞ ( y + b ) e y dy = Z 0 -∞ yde y + be y | 0 -∞ = b - 1 Therefore b MOM = E [ X ] + 1 = 1 n n X i =1 X i + 1 . (b) E [ b MOM ] = 1 n n X i =1 E [ X i ] + 1 = n n ( b - 1) + 1 = b E [ X 2 ] = Z b -∞ x 2 e x - b dx = e - b Z b -∞ x 2 e x dx = b 2 - 2 b + 2 V ar ( b MOM ) = V ar 1 n n X i =1 X i + 1 ! = 1 n 2 n X i =1 V ar ( X i ) = n n 2 · ( b 2 - 2 b + 2 - ( b - 1) 2 ) = 1 n . The estimator is unbiased. (c) Noting that the density of X i is f X i ( x ) = e x - b I { x b } : b MLE = argmax b f ( x 1 ,...,x n | b ) = argmax b n Y i =1 f X i ( x i | b ) = argmax b n Y i =1 e x i - b I { x i b } = argmax b I { b max { x 1 ,...,x n }} e n i =1 x i - nb = max { x 1 ,.... x n } (d) P (max { X 1 ,...,X n } ≤ u ) = n Y i =1 P ( X i u ) = e n ( u - b ) f max { X 1 ,...,X n } ( u ) = ne n ( u - b ) E [ b MLE ] = n Z b -∞ ue n ( u - b ) du = b - 1 n E [ b 2 MLE ] = n Z b -∞ u 2 e n ( u - b ) = b 2 - 2 b n + 2 n 2 V ar ( b MLE ) = b 2 - 2 b n + 2 n 2 - ± b - 1 n ² 2 = 1 n 2 . This estimator is biased. The unbiased alternative is: b MLE + 1 n (e) We use the mean square error
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: MSE ( ) = E ( - ) 2 = V ar ( ) + bias ( 2 ) to compare the estimators. MSE ( b MOM ) = 1 n 1 MSE ( b MLE ) = 1 n 2 + b-1 n-b 2 = 2 n 2 MSE b MLE + 1 n = 1 n 2 Since MSE b MLE + 1 n MSE ( b MLE ) MSE ( b MOM ), b MLE + 1 n is the better of the three using this criterion. Also note that A MLE has fundamental aws. Depending on your sample points, the estimator may end up being larger than some of the observed values, which is impossible under the uniform distribution. 2...
View Full Document

This note was uploaded on 07/22/2009 for the course IEOR 165 taught by Professor Shanthikumar during the Summer '08 term at University of California, Berkeley.

Page1 / 2

Quiz_2_Solutions - MSE ( ) = E ( - ) 2 = V ar ( ) + bias (...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online