Does anyone know a method of curve fitting multiple vectors of data for a single function with about 10 fitting parameters ? There are around 6 variables in the function. Also, for each variable, there is a vector of experimental data. The experimental data is derived by varying that single variable and keeping the rest constant.
The difficult part is to maintain consistency and global accuracy of the function as you fit for one variable then move on to the next. Does anyone have any suggestion on how to do this ?
Ive used origin already for this problem, but I have to do the fitting for each vector of data by hand. The results are still substandard for some variables.
DOes origin have a tool which allows for fitting multiple datasets for
Origin's NLSF allows for fitting a user-defined fitting function with multiple independent variables, and it allows for fitting multiple datasets.
These features can be combined. You first need to define a user-defined fitting function. The Help has more information on how to do this. Make sure you change the Form from Expression to Equation.
When assigning the datasets to the variables (Action:Dataset), you will need to check the 'Fit Multiple Datasets' checkbox. Click the Add data button for each dataset that you are fitting to this function.
Non linear curve fit but for multiple datasets
Posted - 03/15/2002 : 9:59:40 PM
