Message-ID: <loom.20110412T030432-189@post.gmane.org>
Date: 2011-04-12T01:06:29Z
From: Ben Bolker
Subject: simple maths question
Georg Ehret <georgehret <at> gmail.com> writes:
>
> Hi Mrs & Ms "R",
> A simple maths question that I am trying to resolve with R: I need to
> calculate the SE from a pvalue and it's beta... How to do this...?
> Thank you very much and best regards!
> Georg Ehret, Geneva, Switzerland.
Without more information, I don't think you can. **If** you are
assuming a Z test (i.e. the thing you are testing against a null
hypotheses H_0=0 is supposed to be normally distributed) then you
know that
p-value = 2*pnorm(abs(beta/SE),lower.tail=FALSE)
[based on a two-tailed test] and you can use qnorm() to
invert this, but you can't separate beta and SE.