Date:         Wed, 12 May 1999 10:17:50 +0100
Reply-To:     tra <tra@PROTEUS.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
Comments:     To: John Ward <john1131@HOTMAIL.COM>
From:         tra <tra@PROTEUS.CO.UK>
Organization: Proteus Molecular Design Ltd
Subject:      Re: Maximum likelihood estimation
Comments: To: SAS-L@LISTSERV.VT.EDU
Content-Type: text/plain; charset=us-ascii

John,
 
If theta is known, then median = theta/sqrt(2).
 
I presume theta is unknown, and you need to estimate from a set of data x_i :
i=1..n
The mle of theta is the smallest possible value, given x:
mle(theta)=max(x)
Hence
mle(median) = max(x)/sqrt(2)
 
regards
 
Tim
 
John Ward wrote:
 
> I have a question, please.
>
> I want to find the maximum likelihood estimate for the median of the
> following
> distribution. I would really appreciate it if you could give me some help.
>
> f(x)=2x/theta**2 ,    0<x<theta
>    =0, otherwise.
>
> Thank you
>
> John Ward
> New Jersey
>
> _______________________________________________________________
> Get Free Email and Do More On The Web. Visit http://www.msn.com
 
--
T R Auton PhD MSc C.Math
Head of Biomedical Statistics
Proteus Molecular Design Ltd
Beechfield House
Lyme Green Business Park
Macclesfield
Cheshire SK11 0JL
UK
email: tra@proteus.co.uk