| T O P I C R E V I E W |
| sifltijr88 |
Posted - 01/29/2013 : 4:56:26 PM
Origin Ver. and Service Release (Select Help-->About Origin): 8.0 SR0
Operating System: Vista
Hi, I have a couple of questions to do with errors on nonlinear fittings.
I have data (counts (y) and time (x)). When I just plot those I get a good fit for exponential decay (ExpDec1 function) and I also get an error for each variable in the exponential decay. reduced r^2 of approx. 0.95
Q1) How does it find this error (does it just use 1/count as the error on y?)
If I add a column that has 1/count and set that as my error on y, the ExpDec1 fit doesn't return any error on the variables and the reduced r^2 is -0.0005 or something ridiculous like that.
Q2) Why doesn't it fit properly for my errors? |
| 3 L A T E S T R E P L I E S (Newest First) |
| snowli |
Posted - 02/04/2013 : 09:58:51 AM
Sorry that there is no such section in 8.0's Help. We must have improved the documentation in this area later.
You can see it online
http://www.originlab.com/www/helponline/Origin/en/UserGuide/The_Fit_Results.html
Thanks, Snow Li
OriginLab Corp.
quote:
Originally posted by sifltijr88
Thanks for your reply.
Do you happen to know the name of the section with the standard error calculations. I tried searching for "the fit results" and couldn't find the section to which you were referring.
Thanks again!
|
| sifltijr88 |
Posted - 02/01/2013 : 8:31:29 PM
Thanks for your reply.
Do you happen to know the name of the section with the standard error calculations. I tried searching for "the fit results" and couldn't find the section to which you were referring.
Thanks again! |
| snowli |
Posted - 01/29/2013 : 5:53:18 PM
I think you mean the Standard eror for parameters you got in result sheet.
The parameter standard errors can give us an idea of the precision of the fitted values. Typically, the magnitude of the standard error values should be lower than the fitted values. If the standard error values are much greater than the fitted values, the fitting model may be overparameterized.
Choose Help: Origin.
You can search "The fit results", it has a section about Parameter Standard errors. You can read how it's calculated.
If you have an error column, we can use different weighting method in the Fit.
E.g. if you have X, Y, YError column, highlight them to plot a graph.
Choose Analysis: Fitting: Nonlinear curve fit, it will pick the yerror data automatically.
You can go to Data Selection node.
Expand
Input Data: Range 1: y: Weight. U can see different weighting method. 1/error is one of them.
Data under it shows which column is used as error.
Thanks, Snow |
|
|
