T O P I C R E V I E W |
ria7777 |
Posted - 04/18/2013 : 09:16:42 AM dear origin team!
I would like to fit several data points with error and direct weighting. However, I get different results than a friend of mine using the same data set a.nd fitting parameters. Even if I fit the data in her graph, I get the same values for intercept and slope, but significantly different values for standard errors. My friend uses origin 8. (both graphs are attached) How could this be? Which values are "true"? Thanks in advance!!
Origin Ver. and Service Release (Select Help-->About Origin): OriginPro 9.0.0 Operating System: Windows 7 http://www.originlab.com/ftp/forum_and_kbase/Images/graphs incl fitting.xlsx |
4 L A T E S T R E P L I E S (Newest First) |
ria7777 |
Posted - 04/18/2013 : 2:24:26 PM Hey!
Thanks a lot for your help! It's a lot clearer now.
Cheers, Maria |
matthew@originlab.com |
Posted - 04/18/2013 : 1:34:30 PM Try reading through these threads:
http://www.originlab.com/forum/topic.asp?TOPIC_ID=6819
http://www.originlab.com/forum/topic.asp?TOPIC_ID=18412
Essentially, reduced Chi-Sqr divided the Chi-Sqr by degrees of Freedom. This means reduced Chi-Sqr values can be compared easily to those of models with a different number of degrees of freedom.
Matthew OriginLab |
ria7777 |
Posted - 04/18/2013 : 12:37:23 PM Hi!
Thanks for answering that quickly! This tick was indeed the "problem". What exactly does it mean, when it is ticked? In your opinion, should I use the higher or the lower values for the standard deviation?
Thanks again!
P.S. I just sent the data per Email. Thanks for the offer!
|
matthew@originlab.com |
Posted - 04/18/2013 : 09:48:44 AM Hi,
I'm unable to open the attachments, would you mind sending them to us at tech@originlab.com
One quick thing to check would be the option to "Use Reduced Chi-Sqr" under the Advanced tab of the Non-Linear Fit dialog. By default this option is checked, un-checking it will not affect the values of the fit, but will affect the standard errors.
Matthew OriginLab |