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:   Lecture 7 We showed that ν ( S − m ( χ )) l ∈ n ∀ ( ν ( S − m ( χ )) 2 ) 1 / 2 ( nI ( χ )) 1 / 2 . Next, let us compute the left hand side. We showed that ν l ∈ ( X 1 χ ) = which implies  that X ν l ∈ = ν l ∈ ( X i χ ) = n  and, therefore, ν ( S − m ( χ )) l ∈ = ν S l ∈ n − m ( χ ) ν l ∈ = ν S l ∈ . n n n Let X = ( X 1 , . . . , X n ) and denote by ( X χ ) = f ( X 1 χ ) . . . f ( X n χ )    the joint p.d.f. (or likelihood) of the sample X 1 , . . . , X n We can rewrite l ∈ in terms n of this joint p.d.f. as n X ∈ ( X χ ) l ∈ = log f ( X i χ ) = log ( X χ ) = . n χ  χ  ( X   χ ) i =1 Therefore, we can write ∈ ( X  χ ) ∈ ( X χ ) ν Sl ∈ = ν S ( X ) = S ( X )  ( X ) dX n ( X χ ) ( X χ ) = S ( X ) ∈ ( X  χ ) dX = χ S ( X ) ( X χ ) dX = ν S ( X ) = m ∈ ( χ ) .  χ Of course, we integrate with respect to all coordinates, i.e. dX = dX 1 . . . dX n . We finally proved that m ∈ ( χ ) ∀ ( ν ( S − m ( χ )) 2 ) 1 / 2 ( nI ( χ )) 1 /...
View
Full
Document
This note was uploaded on 10/11/2009 for the course STATISTICS 18.443 taught by Professor Dmitrypanchenko during the Spring '09 term at MIT.
 Spring '09
 DmitryPanchenko
 Statistics

Click to edit the document details