| T O P I C R E V I E W |
| Chillkroete |
Posted - 01/21/2014 : 5:24:59 PM
Dear Community,
I guess I have a very simple question, but I am completely new to Origin.
I want to do a simple regression fitting with a bunch of Data points. As exponential fitting function I want to use: A *exp(-B*x)
But after plotting my dataset I cannot find a fitting curve like that. Do I have to create a new fitting curve or did I just overlook the right curve?
Thank you for your support!
Cheers
Chillkroete |
| 6 L A T E S T R E P L I E S (Newest First) |
| Chillkroete |
Posted - 01/28/2014 : 10:04:07 AM
Allright, now it works perfectly! Thanks a lot at all! |
| Echo_Chu |
Posted - 01/26/2014 : 01:15:26 AM
Hi, Chillkroete
It looks you did not hit the Fit until converged button when using the ExpDecay1 function. Please look at the fitted curve after I hit the Fit until converge button. However, there is also warning message to remind "you may have overparameterized the fitting function. Fixing one of them may eliminate this problem"
So I also tried fix x0 to 2010 and the fit looks good.
 |
| Chillkroete |
Posted - 01/25/2014 : 04:38:50 AM
Cool, that works quite well for most of my Data, thank you!
But I have one Dataset, which does not look nice:
Do you have an idea what might be the preoblem, why the fit looks weired? Looks like the points in te middle are not considered.
Thank you for your reply! |
| lkb0221 |
Posted - 01/23/2014 : 4:48:33 PM
Hi#65292; Micha
I think you can try ExpDecay1 function (y0 + A1 * exp(-(x-x0)/t1)).
Still not a good fit, but I think it's acceptable.
It is the same as Exp2pMOD1 but with a y offset.
Zheng
OriginLab
|
| Chillkroete |
Posted - 01/22/2014 : 6:04:25 PM
Hi Amanda,
thank you for your reply!
I tried EXP2PMOD1 and I found the correct formula (of course you were right! ;) )
But unfortunately the fitting looks absolutely weired, as you can see here:
On the sceenshoot you only see the fit very light. Its just a line at the X-Axis.
If I use Asymptotic1 for example It looks quite well, but of course It does not have the desired parameter like: A *exp(-B*x)
Do you have an Idea what I did wrong?
Thank you for your help! |
| AmandaLu |
Posted - 01/22/2014 : 05:55:44 AM
Hi Chillkroete,
You can fit with a build-in function Exp2PMod1, under Exponential category. Here is a video tutorial that can help you quickly get started with Origin Nonlinear curve fitting dialog:
http://originlab.com/Index.aspx?go=Support/VideoTutorials&pid=1564
Sincerely,
Amanda |
