# Deriving Beta-distribution from F-distribution

• February 15th 2010, 01:46 PM
epsilonbrain
Deriving Beta-distribution from F-distribution
I've tried searching everywhere (books, internet) on clues how to prove this

If X~ F(m, n) then $W=\frac{m\frac{X}{n}}{1+{m\frac{X}{n}}}$ has a beta density with a = m/2 and b=n/2

I'd appreciate it if someone points me to the right direction on where to start because that's basically my problem right now. Thanks ahead.
• February 15th 2010, 02:53 PM
matheagle
I would just use calculus one.

$f_W(w)=f_X(x)\left|{dx\over dw}\right|$