offset for lognormal prior

651 views
Skip to first unread message

MD

unread,
Feb 8, 2010, 2:06:04 PM2/8/10
to beast-users
Hi beast-users,

I am relatively new to Beast and have a very simple question about
lognormal priors.

I want to put a lognormal prior distribution on the root of my tree: I
have both a minimum age (~2 ma, from the fossil record) and an
estimate of mean divergence time (~2.83 ma. from previous
literature). If I set the offset to 2.0, should I then designate my
mean (in real space) as 0.83? Considering the posterior distributions
(for root age) that I get when I run Beast under this scenario, I
think this is correct.

I apologize if this question has been answered previously (or is too
simple to ask here). I did search the group for this topic, but could
not find a satisfactory answer. For all other beast-related
questions, this forum has been extremely helpful! Thank you!

MD

Luis Delaye

unread,
Sep 9, 2015, 9:32:57 AM9/9/15
to beast-users

Hi MD,

By looking at the tutorial:


page 23, it seems that you have to add an offset of 2 ma and then to add a Mean of 0.83. 

However I'm also new to BEAST. Hopefully somebody else contributes to this question. I'm also figuring out how to use offset properly.

Best!

L
Reply all
Reply to author
Forward
0 new messages