| T O P I C R E V I E W |
| maryam62 |
Posted - 02/10/2021 : 05:24:29 AM
Hi
We used a nonlinear surface fit, and Poly2D Function, to fit our data. In the results section, figure below:
Is reduced chi square an Absolute error or relative error?
|
| 6 L A T E S T R E P L I E S (Newest First) |
| maryam62 |
Posted - 02/11/2021 : 06:23:06 AM
Dear Yiming Chen
Thanks your response. |
| YimingChen |
Posted - 02/10/2021 : 5:07:48 PM
If you check equation (19) on that help page, the residual is calculated by yfit-y, so I guess it's absolute error? Is this what you mean?
James |
| maryam62 |
Posted - 02/10/2021 : 4:01:39 PM
Can the value we report for this quantity (the reduced chi-square value) be an absolute or relative error, or is it not an error at all? |
| YimingChen |
Posted - 02/10/2021 : 3:39:26 PM
Hi,
What do you mean by absolute/relative/none here? were you asking whether the reduced chi-square value changes if data scales?
James |
| maryam62 |
Posted - 02/10/2021 : 1:18:33 PM
Dear Yiming Chen
Thanks your response.
I read your suggested link. However, I could not conclude whether the reduced chi-square value was absolute or relative or none at all. |
| YimingChen |
Posted - 02/10/2021 : 09:24:16 AM
It's defined as the residual sum of squared divided by the freedom. See page below:
https://www.originlab.com/doc/Origin-Help/NLFit-Theory#Reduced_Chi-Sqr
James |
