-------- Original Message --------
Subject: RE: bimodal variable
Date: Sun, 26 Sep 2010 21:05:26 -0400
From: F. James Rohlf <[email protected]>
Reply-To: [email protected]
Organization: Stony Brook University
To: [email protected]
That is not a mixture of two uniform distributions! Perhaps the
simplest way to get random numbers corresponding to a bimodal
distribution is to use a uniform 0-1 random number to choose among two
distributions (e.g., < .5 choose one distribution and > 0.5 choose the
other) and then either draw a number from one distribution or from the
other. The distributions you are drawing from need to have modes. The
distribution that you show could perhaps be simulated by a mixture of a
binomial with p around 0.1 and another with a p around 0.9. One could
probably do better if you had some theory about what types of
distributions would correspond to the process you are simulating.
----------------------
F. James Rohlf, Distinguished Professor
Dept. Ecology and Evolution, Stony Brook University, NY 11794-5245
Please consider the environment before printing this email
-----Original Message-----
From: morphmet [mailto:[email protected]]
Sent: Sunday, September 26, 2010 1:01 PM
To: morphmet
Subject: Re: bimodal variable
-------- Original Message --------
Subject: Re: bimodal variable
Date: Sun, 26 Sep 2010 12:45:02 -0400
From: Samor Gandhi <[email protected]>
To: [email protected]
Thanks for reply. A variable between zero and one (mixture of two
uniformly distribution may be) similar to the attached graph!
Regards,
Samor
--- On *Sun, 26/9/10, morphmet
/<[email protected]>/*
wrote:
From: morphmet <[email protected]>
Subject: Re: bimodal variable
To: "morphmet" <[email protected]>
Date: Sunday, 26 September, 2010, 22:07
-------- Original Message --------
Subject: Re: bimodal variable
Date: Sun, 26 Sep 2010 12:20:27 -0400
From: F. James Rohlf <[email protected]
</mc/[email protected]>>
Reply-To: [email protected]
</mc/[email protected]>
To: Morphmet <[email protected]
</mc/[email protected]>>
What distribution? A mixture of two binomial distributions?
------Original Message------
From: morphmet
To: Morphmet
ReplyTo: Morphmet
Subject: bimodal variable
Sent: Sep 26, 2010 12:05 PM
-------- Original Message --------
Subject: bimodal variable
Date: Sat, 25 Sep 2010 20:17:07 -0400
From: Samor Gandhi <[email protected]
</mc/[email protected]>>
To: [email protected]
</mc/[email protected]>
Hello,
I would like to simulate bimodal variable between zero and one,
where
more data are heaped to zero and one and less between these. I am
very
thankful to nay help!
Regards,
Samor
-------
Sent remotely by F. James Rohlf
