[R] fitting a curve to data points

[Dieter Menne]

<carlos.grohmann <at> gmail.com> writes:

> 
> Hello all. This is likely to be a silly question, but I have a set of
> data points and I want to fit a curve to it, like this:
> http://www.igc.usp.br/pessoais/guano/temp/curve.png

While you could use a loess-curve, interpreting the rising branch a the 
end is probably close to cheating. I would simply guess that this is an
exp(-t/t0) curve with a few outliers. You should really have good reasons 
with so few datapoints in the upward branch.

Dieter