| T O P I C R E V I E W |
| mpillana |
Posted - 05/01/2021 : 07:34:55 AM
Origin Ver. and Service Release (Select Help-->About Origin):
Operating System:
Origin pro 2019 (64 bits ) 9.6.0.172 (academic)
windows 7
Hi
(1) - I am performing an "average multiple curves", with the same number of points as the input data.
- I got the "Std Dev" and the "Std Err values reported".
- At first, I thought "Std Dev" should be the one being plot as error bars in the graph. Now I am doubting if "Std Err" would be better (I think I am aware that StdErr is not the Standard error of the mean, and I guess that this is related to the goodness of the average fitting? ...). Any opinions on this?
(2) - Then I am getting a "Boltzman" fit of the average curve, and reporting the values of the parameter I am interested in (x0) + the "Standard Error" for that parameter, which I read defines the goodness of the fit... Should I better report other errors such as the t-value or the prob for the parameter? Or even global "errors" such as the Rsquare or whatever?
Thanks a lot!!
Marta
Marta |
| 1 L A T E S T R E P L I E S (Newest First) |
| YimingChen |
Posted - 05/03/2021 : 10:08:09 AM
Hi,
(1) Std Err is the Standard error of the mean. Regarding using which as error bar, here is a nice article:
https://blogs.sas.com/content/iml/2019/10/09/statistic-error-bars-mean.html
(2) In terms of fitting, R-squared is better to be included in the output. If you would like to determine if any parameters is significantly different from 0, then show the P-value for that parameter.
James
|
|
|
