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