| Author |
 Topic  |
|
|
kornster
Finland
Posts |
 Posted - 05/04/2006 : 12:13:12 PM
|
Origin Version (Select Help-->About Origin): 6.1
Operating System: win xp
Hi! I need to fit a dataset consisting of 3 independent variables (concentrations of different compounds) and one variable dependent on these three independent variables (rate of reaction in the presence of different amounts of the aforementioned compounds) into a user-defined equation. I used to do this for two independent and one dependent variable by just fitting x,y,z columns using a simple equation formulated to this effect:
z=(P1+P2*(x/P3))/(1+(x/P3)+(y/P4))
Now, I want to fit 3 independent variables into a rather simple equation which has a parameter P5 and another independent variable in addition to the ones in the equation above. Apparently fitting data in 4D is quite difficult, so is there another way or can origin handle the fitting of 4D data?
Thanks a lot. |
|
|
easwar
USA
1965 Posts |
Posted - 05/04/2006 : 12:21:43 PM
|
Hi kornster,
The fitter can handle more than 2 indep vars...you can simply type in any number of indep vars and dep vars and then do approriate dataset assignments for each. The minimizing routine will work with all those variables. When doing this manually, a graph pops up, and what is plotted there and the fit line shown there may not make sense for more than one indep var case (because for more than one indep var, cannot represent data and fit in a 2D plot), but internally the fitting will work just fine and the parameters you get from the fit result should be all fine.
Did you try assiging more than 2 indep vars and fit with it? If you run into problems, please send data and your fitting function to tech support.
Easwar
OriginLab
|
 |
|
| |
Topic |
|
|
|
fitting 4D data with NLSF