# post9 - Likelihood Ratio Tests Composite versus Composite...

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

Likelihood Ratio Tests Composite versus Composite H 0 : θ Θ 0 vs. H 1 : θ Θ 1 Ω=Θ 0 Θ 1 Extend the idea of N-P Lemma, which is based on the likelihood ratio L ( θ 0 ) L ( θ 00 ) . Likelihood ratio statistic λ ( x 1 , ··· ,x n )= sup θ Θ 0 L ( θ ) sup θ Ω L ( θ ) C = { ( x 1 , ··· ,x n ):0 λ ( x 1 , ··· ,x n ) λ 0 } where λ 0 is chosen so that Pr ( λ ( x 1 , ··· ,x n ) λ 0 ; H 0 )= α. 1 Example : X N ( θ 1 2 ) , Ω= { ( θ 1 2 ): -∞ < θ 1 < , 0 2 < ∞} H 0 : θ 1 =0vs. H 1 : θ 1 6 =0 Θ 0 def = { ( θ 1 2 ): θ 1 =0 , 0 2 < ∞} ⊂ Ω. Then H 0 :( θ 1 2 ) Θ 0 vs. H 1 : not H 0 L 0 )= L (0 2 ; x 1 , ··· ,x n ) L (Ω) = L ( θ 1 2 ; x 1 , ··· ,x n ) L ( ˆ Θ 0 )= L ( ˆ Ω) = 2 λ = L ( ˆ Θ 0 ) L ( ˆ Ω) =( 1 1+ n ¯ x 2 ( x i - ¯ x ) 2 ) n/ 2 λ 0 λ λ 0 n | ¯ x | q ( x i - ¯ x ) 2 / ( n - 1) c C = { n | ¯ x | q ( x i - ¯ x ) 2

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 09/10/2009 for the course STATS 517 taught by Professor Song during the Fall '07 term at Purdue.

### Page1 / 3

post9 - Likelihood Ratio Tests Composite versus Composite...

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

View Full Document
Ask a homework question - tutors are online