29.07.2014 Views

Asymptotic Methods in Statistical Inference - Statistics Centre

Asymptotic Methods in Statistical Inference - Statistics Centre

Asymptotic Methods in Statistical Inference - Statistics Centre

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

75<br />

To see this take 0 = 0 for simplicity. Let be the<br />

( 1 1) d.f. and the ( 2 1) d.f. Let =(1−<br />

)+ for ∈ [0 1]; require 1 and 2 to satisfy<br />

(1 − ) 1 + 2 =0 (8.1)<br />

Then ∈ F. Represent the rejection region as ‘X ∈<br />

’, then the level is<br />

( ) = (X ∈ )<br />

=<br />

≥<br />

Z<br />

Y<br />

<br />

=1<br />

(1 − ) Z <br />

{(1 − )( )+( )} 1 ··· <br />

Y<br />

=1<br />

( ) 1 ··· <br />

= (1− ) (X ∈ )<br />

= (1− ) ³√<br />

¯ ≥ ´<br />

<br />

For any this may be made arbitrarily near 1 by choos<strong>in</strong>g<br />

sufficiently small, 1 sufficiently large (how?),<br />

and 2 = −(1 − ) 1 to satisfy (8.1).<br />

That <strong>in</strong>f ( ) = 0 may be shown similarly, by replac<strong>in</strong>g<br />

( )by1− ( )and by its complement,<br />

and then proceed<strong>in</strong>g as above to obta<strong>in</strong><br />

1 − ( ) ≥ (1 − ) ³ √<br />

¯ ´<br />

<br />

(Thanks to J. Sheahan for this second part.)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!