hwk3_soln

# hwk3_soln - X i 2 n Î âˆ’ 2 log Î n X i =1 X i 2 n X i =1...

This preview shows page 1. Sign up to view the full content.

IEOR E4702 Statistical Inference for Financial Engineering HWK Solution 3 1. Under H 0 the likelihood is given by L H 0 ( λ 0 )= n Y i =1 f ( X i | λ 0 )= n Y i =1 e λ 0 ( λ 0 ) X i X i ! , and log L H 0 ( λ 0 )= n λ 0 +log( λ 0 ) n X i =1 X i n X i =1 log( X i !) . Without constraints under H a ,wehave log L H a ( λ )= n λ +log( λ ) n X i =1 X i n X i =1 log( X i !) . Taking the derivative with respect to λ and setting it to zero we have n + P n i =1 X i λ =0 . Therefore, the MLE under H a is ˆ λ = P n i =1 X i n = ¯ X. Thus, log L H a ( ˆ λ )= n ¯ X +log( ¯ X ) n X i =1 X i n X i =1 log( X i !) . Therefore, the test statistic is given by Λ =2 l o g ( L H a ( ˆ λ ) L H 0 ( λ 0 ) )=2log( L H a ( ˆ λ )) 2log( L H 0 ( λ 0 )) = 2 n ¯ X +2log( ¯ X ) n X i =1 X i 2 n X
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: X i !) + 2 n Î» âˆ’ 2 log( Î» ) n X i =1 X i + 2 n X i =1 log( X i !) = 2 n ( Î» âˆ’ Â¯ X ) + 2 Â¡ log( Â¯ X ) âˆ’ log( Î» ) Â¢ n X i =1 X i , which asymptotically has Ï‡ 2 distribution with d.f. one. Thus, we reject the null hypothesis if 2 n ( Î» âˆ’ Â¯ X ) + 2 Â¡ log( Â¯ X ) âˆ’ log( Î» ) Â¢ n X i =1 X i . â‰¥ Ï‡ 2 Î± , 1 . 2. See the attached excel f le. 1...
View Full Document

## This note was uploaded on 10/18/2010 for the course IEOR 4702 taught by Professor Kou during the Spring '10 term at Columbia.

Ask a homework question - tutors are online