Message-ID: <B2C608BD-E266-464C-A2F4-CABCD99C1F84@comcast.net>
Date: 2012-08-22T17:37:50Z
From: David Winsemius
Subject: log-normal distribution fitting with expected value = 1
In-Reply-To: <CAGs0_Qqrc2bAB95sP1aMy4dZVfe0t59BB89pOFzHUuyG4=AE6Q@mail.gmail.com>
On Aug 22, 2012, at 2:23 AM, Biophil wrote:
> Dear R users,
> I would like to estimate mu and sigma of a log-normal distribution,
> where I
> know that the expected value is 1, as it is a normalized distribution.
Are you sure that is what "normalized distribution" actually means in
the context you are reading it? I would have assumed that it meant
that the integral over its domain equaled 1, which is quite different
than its mean equaling 1.
> That
> means as E(x) = exp (mu + 1/2*sigma^2) = 1 that 2*mu = -sigma^2 .
> Therefore
> I only need to fit one parameter either sigma or mu. How could I do
> this in
> R?
> Thank you very much for your help!
> biophil
>
--
David Winsemius, MD
Alameda, CA, USA