{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# post7 - Best Critical Region Certain Best Tests Methods of...

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

Certain Best Tests Methods of constructing good tests simple H 0 : θ = θ 0 vs. simple H 1 : θ = θ 00 Best tests Best critical regions X 1 , · · · , X n : random sample from f ( x ; θ ) What is the best critical region? 1 Best Critical Region C is a subset of A . Then C is called a best critical region of size α for testing H 0 : θ = θ 0 against H 1 : θ = θ 00 if, for every subset A ∈ A for which Pr [( X 1 , · · · , X n ) A ; H 0 ] = α, (a) Pr [( X 1 , · · · , X n ) C ; H 0 ] = α, (b) Pr [( X 1 , · · · , X n ) C ; H 1 ] Pr [( X 1 , · · · , X n ) A ; H 1 ] 2 Test Statistics: Likelihood Ratios Test Statistic: a function (statistic) of the sample data on which the decision is to be based. Likelihood ratio statistic: L ( θ 0 ) L ( θ 00 ) If this ratio is “small”, then we reject H 0 . How “small” is small? 3 Neyman-Pearson Theorem (N-P Lemma) X 1 , · · · , X n is a random sample from f ( x ; θ ). L ( θ ; x 1 , · · · , x n ) = f ( x 1 ; θ ) f ( x 2 ; θ ) · · · f ( x n ; θ ) Ω = { θ 0 , θ 00 } and k > 0 Let C be a subset of A such that (a) L ( θ 0 ; x 1 , ··· ,x n ) L ( θ 00 ; x 1 , ··· ,x n ) k , ( x 1 ,

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.
• Fall '07
• Song
• Null hypothesis, Statistical hypothesis testing, Statistical theory, Likelihood principle, Uniformly most powerful test

{[ snackBarMessage ]}