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

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

173<br />

• Assume:<br />

(C1) Identifiability: 1 6= 2 ⇒ the distributions<br />

1 2 are dist<strong>in</strong>ct. (This is violated if, e.g.,<br />

1 ∼ ( 1) but only = || is observed,<br />

then ± both lead to the same distributions<br />

of : () =[ ( − )+ ( + )] ( <br />

0).)<br />

(C2) The parameter space is an open <strong>in</strong>terval:<br />

Ω = ³ ¯´<br />

(possibly <strong>in</strong>f<strong>in</strong>ite). (Ensures that maxima<br />

of the likelihood function correspond to critical<br />

po<strong>in</strong>ts, assum<strong>in</strong>g differentiability.)<br />

(C3) The observations 1 arei.i.d.with<br />

density or p.m.f. (). (Then () = P log ( ).)<br />

(C4) The support = { | () 0} is <strong>in</strong>dependent<br />

of . (This excludes cases like ∼<br />

(0), which can be handled us<strong>in</strong>g other techniques.)<br />

(C5) For all ∈ , () isdifferentiable w.r.t. ,<br />

with derivative 0 (). (Then maxima of satisfy<br />

the likelihood equation 0 () = P 0( )<br />

( ) =0.)

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

Saved successfully!

Ooh no, something went wrong!