It's hard to say without seeing the data. It could be the data, it could
be the starting values, it could be the model choice.
Jean
On Wed, Jun 19, 2013 at 10:45 AM, pakoun wrote:
> Yes it should look like that... what i am doing is a variogram fit . But
> the
> data of course are spread almo
Yes it should look like that... what i am doing is a variogram fit . But the
data of course are spread almost all over.. I would guess might be problem
with the data only?
--
View this message in context:
http://r.789695.n4.nabble.com/nls-singular-gradient-as-always-tp4669859p4669898.html
Sent
What does a plot of your data look like?
plot(ndat$dist, ndat$vario.dNEE)
Anything remotely like a two-parameter single exponential rise to a maximum
from zero as shown on this webpage, for example?
http://www.graphpad.com/guides/prism/6/curve-fitting/index.htm?reg_classic_1assoc.htm
Jean
O
Hi all. Sorry for posting again such a topic but I went through previous
posts but couldn't find a solution.
I use the following code to fit an exponential model to my data. I have 4
different datasets. For 3 datasets nls seems to work fine and I have no
error messages. But for 1 dataset I am get
4 matches
Mail list logo
