I am wondering ho to perform a non-linear curve fitting considering the uncertainties of experimental data points. I am an R user (not advanced user if I am honest) and I know how to fit a curve using non-linear fitting. However I have no idea how to do the same but including the individual uncertainties of experimental data points. Any ideas?
Asked
Active
Viewed 80 times
1
-
You can assign weights to the individual data points that represent this uncertainty. Check my answer [here](http://stackoverflow.com/questions/31837610/forcing-nls-to-fit-a-curve-passing-through-a-specified-point/31888068#31888068) how it can be done. The vector `we` in this example just needs to be replaced by your actual weights. – Cleb Apr 01 '16 at 11:26
-
Thanks for your answer Cleb. My question now is whether it would be possible to include the uncertainties in the case of linear fit. Is that feasible? – user3329055 Apr 06 '16 at 05:35
-
Sure, at least I do not see any reason why it would not work. You should be able to easily adapt the code from the link above to your problem. If you need help with that, then please provide some data and code as well as your desired model etc :) – Cleb Apr 06 '16 at 10:41