Is it possible to add fitting lines pertaining to different curve fits in a single plot in Origin? If yes, how can it be done? I have searched a lot on internet in this regard, but could not get any satisfactory answer. I have attached the graph showing different lines for different sections of the plot. I am able to do the curve fragmentation, but not able to put lines for different fits.
In the attached file, the dashed and solid lines are fits for different powers of x.
* This graph is taken from a published paper just to clarify my case and is not my data.
Thanks.
Hi Poonam,
Perhaps you should look at Regression Discontinuity Design. The idea is to introduce discontinuity between the fitting models.
Hi Poonam,
maybe this will be useful
http://www.originlab.com/fileExchange/details.aspx?fid=325
Dear Poonam,
Try to collect all data and apply fitting curve for all at once.
What do you mean by "single plot in origin"?
Clarify your request; you may get better answers.
Best regards
Thank you all for the responses.
Dear Issam,
Actually, I want to add different trendlines for a single data series. Similar to the one that is shown in the image that I had shared earlier. In this graph, the dashed lines are for a linear fit, while the solid line is a 1/2 fit for the x-axis. I have tried fitting different portions of the graph differently, in origin.
I tried using power function in origin to obtain the trendline for 1/2 fit, but was not able to do so.
I am using OriginPro 8.
Hi Poonam,
Correct me if I'm wrong. Your task objective is to find the right nonlinear model to fit the curve generated by the data that exhibit an exponentially decaying trend. However, you have not found a suitable model yet. So you thought one way to solve this is by resorting to piecewise curve fitting method (a.k.a. Spline interpolation).
If the data display a continuous trend without noise, most of the time you can fit the data reliably over a range with one or several suitable models. Based on the continuous data trend in the figure, you may try using an exponential model, or a rational polynomial model, or a combination of both models like the Planck's law. Furthermore, if you know the physics that explains the behavior of the data, then you should look for the standard model that underlies this particular case.
@Poonam,
[Origin is an industry-leading scientific graphing and data analysis software.]
It is new to me; I am familiar with Maple, SW, Mathematica, Math lab I think your problem is technical and not a mathematical one.
My advice to read "Origin" help and study the examples there, you may find a solution : ( How to make a thick graph for your data ),
OR contact the home page of "Origin Lab" through help center in the Support category.
https://www.originlab.com/
Regards
Hi Poonam,
There is a tool called Fitting Function Builder which allows the user to fit multiple models at different segments of a dataset.
https://www.originlab.com/doc/Origin-Help/FitFunc-Builder
Dear Yew-Chung Chak,
Thanks. The link is quite useful. My data is quite similar to the one shown in the graph. I have added separate segments to fit it.
@Victor Sir,
Thank you for sharing the link. I have broken the plot in different segments to fit it.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs