Skip to content

error message

4 messages · forkusam, Peter Dalgaard, Uwe Ligges

#
Hi,
Can someone please tell me what such an error message
could mean. i.e where a problem must have arised.

Error in uniroot(function(n) eval(p.body) - power,
c(2, 1e+07)) : 
        f() values at end points not of opposite sign


Thank you
cilver

=====
=====================
Sylvie B. Forkusam
Eppelheimer Str.52/A2-5-2
69115 Heidelberg, Germany
Tel: (0049)-06221/346913
Mobile: 0179-6816276
#
forkusam <forkusam at yahoo.com> writes:
Looks like the innards of one of the power calculations, finding n to
achieve a given power. You'd likely get an error like that if the
power is not between 0 and 1, but how about telling us what you were
trying to do?
2 days later
#
I am trying to calculate n, using power.t.test

power.t.test(n=NULL,delta=delta, sd=sigmaEins,
sig.level= alpha, power=power, type=c("two.sample"),
alternative=c("two.sided"))

The values of the parameter are read in from a file.
 I have no idea what the error message means.
--- Peter Dalgaard <p.dalgaard at biostat.ku.dk> wrote:
=====
=====================
Sylvie B. Forkusam
Eppelheimer Str.52/A2-5-2
69115 Heidelberg, Germany
Tel: (0049)-06221/346913
Mobile: 0179-6816276
#
On Sat, 24 Jan 2004, forkusam wrote:

            
Peter Dalgaard asked you to specify a *reproducible* example.
So, please specify some values for delta, sigmaEins, alpha, power.

Uwe Ligges