| T O P I C R E V I E W |
| xzw102 |
Posted - 04/06/2005 : 5:28:27 PM
In nonlinear curve fitting, if I choose "Statistical" in Weighting Method, why the fitting errors changes a lot depending on whether I check the "Scale Errors with Standard Deviation" or not?
If I want to get the ratio (like Mean +/- SD) of parameters from two fittings of different curves, Should I check it or not? |
| 1 L A T E S T R E P L I E S (Newest First) |
| Leo_Li |
Posted - 04/07/2005 : 01:14:50 AM
Dear Wang,
If "Scale Errors with Sqrt(reduced chi^2)" Check Box is selcted, Origin will multiply the reported error by the square root of the reduced chi-squared. This is why the fitting errors change.
You can find more information in Origih Help:
Analysis Curve Fitting | The Nonlinear Least Square Fitter | Starting the Fitting | Controlling the Fitting Procedure
Regarding how to present your fitting parameters (Mean +/- SE), I am not very sure whether the above option should be checked. But It seems better to uncheck it since you already have error bars for your data (i.e. statistical weighting). |
|
|
