Curve fitting

[SardarM]

Hello everyone,

Does anyone has any idea about fitting a curve based on Error Function? I can fit my data in Matlab based on erf but I could not find any command in LabVIEW for this kind of fitting.

I guess I should use "Mathscript RT Module" but if anyone has an idea to solve it without Mathscript, it would be much appreciated.

[mikeporter]

http://www.ni.com/white-paper/6954/en

This seems to have a lot of information, is this what you are looking for?

Mike...

[johnsold · Accepted solution]

An error function VI is in the Mathematics palette, so you should be able to build a model VI which can be used with the Lev-Mar fitting VI.

Lynn

[locktofer]

And how exactly would you do that? Seems very complicated, any help you can offer, please advise.

Regards,

[johnsold]

This thread is three years old and has been marked as solved. 

You will probably get better help if you start a new thread. Mention that you have read this one. Then be as specific as possible about what you are trying to do, what you have attempted, what errors you get, where your data comes from adn so on. The more information you provide and the more specific your questions, the better the answers you will get.

Lynn

[altenbach]

---

@locktofer wrote:

And how exactly would you do that? Seems very complicated, any help you can offer, please advise.

---

Sometimes searching the forum before posting is easiest. 🙂

You might have found this solution. Try it!

[locktofer]

Thank you both for your responses. However, this is not quite what I am trying to do.

I am trying to curve fit the waterfall curve for the bit error ratio of digital signals. I have a bunch of data that I know approximately follows the complimentary error function, I need to fit it to this function. How do I do it. Altebach, the function that you use here in your example doesnt look anything like the waterfall curve, so I am not sure that I can use that. The previous person wants me to start a new thread, maybe I will do that as well.

Thanks

[altenbach]

The method is independent of the model function. If you have a mathematical model with a few fittable paramters that describes your data,  you can use the method. All you need to do is change the model accordingly.

If you want more detailed help, attach your data and a link to a web page describing the theory. How many dimensions do you have for the independent variable? 

[locktofer]

Ok, again thanks for the reply. I will give it a try with some actual data and let you know how it worked. Thanks again for your time in looking at this.

Best regards,