| T O P I C R E V I E W |
| leejuyeon |
Posted - 06/09/2022 : 9:55:15 PM
Hello
I have the following problem:
z=a+bx+cy
I have a xyz-dataset and i need to find a,b,c parameter.
I have no idea to solve this problem.
Sould I have to do 'nonlinear implicit curve fit'? Or should I have to do 'multiple linear regression'?
Then, should I need to set my 'x, y, z-dataset' as 'x1, x2, y-data' ???? Or should I need to set my 'x, y, z-dataset' as 'x, y, z-data' ????
should I need to set function type as 'equations : f=a+bx+cy-z' ???
Or should I need to set fuction type as 'LabTalk Scrip'??? But I have no idea to make a script.
Also, I followed this post and I tried a lot AND failed. (POST: https://my.originlab.com/forum/topic.asp?TOPIC_ID=1734 )
Please help me. This problem is killing me. Please... Help me ASAP...
Also, if you have something misunderstand, I wil explain.
Thank you so much. Have a good day.
leejuyeon |
| 3 L A T E S T R E P L I E S (Newest First) |
| Shirley_GZ |
Posted - 06/10/2022 : 05:42:55 AM
Hi leejuyeon,
Let me take back my previous suggestion.
I think that the best way to get the parameter values is to perfrom multiple linear regression on your data.
https://www.originlab.com/doc/Origin-Help/Multi-Regression-Dialog
https://www.originlab.com/doc/Quick-Help/Multiple-Linear-Regression
Thanks,
Shirley
Originlab Technical Service Team |
| leejuyeon |
Posted - 06/10/2022 : 12:23:07 AM
Hi, Shirley Tian
Thank you for your reply.
I understood about nonlinear cureve fit.Thank you so much.
Well, I am wondering about..... what kind of 'FUNCTION TYPE' to choose.
Can you answer me what kind of 'FUCTION TYPE' I should choose?
After that, how do I set up the equation (or script or sth else)?
I'm in a bit of a hurry. Please answer me ASAP.
Thank you.
leejuyeon
quote:
Originally posted by leejuyeon
Hello
I have the following problem:
z=a+bx+cy
I have a xyz-dataset and i need to find a,b,c parameter.
I have no idea to solve this problem.
Sould I have to do 'nonlinear implicit curve fit'? Or should I have to do 'multiple linear regression'?
Then, should I need to set my 'x, y, z-dataset' as 'x1, x2, y-data' ???? Or should I need to set my 'x, y, z-dataset' as 'x, y, z-data' ????
should I need to set function type as 'equations : f=a+bx+cy-z' ???
Or should I need to set fuction type as 'LabTalk Scrip'??? But I have no idea to make a script.
Also, I followed this post and I tried a lot AND failed. (POST: https://my.originlab.com/forum/topic.asp?TOPIC_ID=1734 )
Please help me. This problem is killing me. Please... Help me ASAP...
Also, if you have something misunderstand, I wil explain.
Thank you so much. Have a good day.
leejuyeon
leejuyeon |
| Shirley_GZ |
Posted - 06/09/2022 : 10:22:49 PM
Hi leejuyeon,
In this case, you can consider to define the function with two independent variables VS. one dependent variable.
Please refer to the tutorials below:
https://www.originlab.com/doc/Tutorials/Fitting-MultiIndepVar
https://www.originlab.com/doc/Tutorials/Fitting-Datasets
Thanks,
Shirley
|
|
|
