| T O P I C R E V I E W |
| milybh |
Posted - 03/31/2012 : 08:07:21 AM
Hello:
Need help in fixing the following problem:
I have generated a histogram from my dataset and after analyzing it using 'peak analyzer'and fitting the peaks, I get the maximum peak height much beyond the maximum counts in the respective histogram e.g. if the max. count (histogram)= 7, the max. height of the fitted peak = 18, though the peak is centered around the maximum count.
Is there any way to modify the peak height parameter prior to fitting so that the final peak height is closer to the max no. of counts?
Also, I would like to know what algorithm does the "fit peaks" function of the "peak analyzer" uses.
Thanks in advance.
Mily.
|
| 5 L A T E S T R E P L I E S (Newest First) |
| milybh |
Posted - 04/05/2012 : 05:59:58 AM
Sorry Easwar, instead of sending the whole project, i had sent the data worksheet only. I am sending the file once again with complete analysis and thank you for your response regarding the algorithm used in peak-fitting.
Mily.
quote:
Originally posted by easwar
Hi Mily,
We replied, asking for your project that includes your PA results and your histogram, what you sent was just data in one sheet. So please send the OPJ with all of your results.
As for algorithm, Peak Fitting with Peak Analyzer uses the Levenberg-Marquardt iterative algorithm to fit the data/optimize the peak parameters. This is the same algorithm used in NLFit nonlinear curve fitting tool:
http://originlab.com/www/helponline/Origin/en/Category/Theory_of_Nonlinear_Curve_Fitting.html
Easwar
OriginLab
|
| easwar |
Posted - 04/04/2012 : 6:49:32 PM
Hi Mily,
We replied, asking for your project that includes your PA results and your histogram, what you sent was just data in one sheet. So please send the OPJ with all of your results.
As for algorithm, Peak Fitting with Peak Analyzer uses the Levenberg-Marquardt iterative algorithm to fit the data/optimize the peak parameters. This is the same algorithm used in NLFit nonlinear curve fitting tool:
http://originlab.com/www/helponline/Origin/en/Category/Theory_of_Nonlinear_Curve_Fitting.html
Easwar
OriginLab
|
| milybh |
Posted - 04/04/2012 : 03:16:16 AM
Hi,
Just now, sent the file to tech support. I would also like to add that I am using a bin size of 1.
Also, it would be great if you could let me know which algorithm is used for curve fitting.
Thanks again.
quote:
Originally posted by milybh
Hello Easwar,
Thank you for your reply. I just figured out that if I decrease the bin size in the "frequency count" column and keep it at a value identical to the one I use in the histogram, the curve height is okay. However, now there's a new problem...the curve centre seems shifted at times compared to the histogram.
I am sending the file as per your suggestion.
Mily.
quote:
Originally posted by easwar
Hi Mily,
Please send the Origin project to tech support so we can take a look. Use the "Send File to Tech support" link on top right of this page, and refer to this post in your e-mail.
Easwar
OriginLab
|
| milybh |
Posted - 04/04/2012 : 03:00:04 AM
Hello Easwar,
Thank you for your reply. I just figured out that if I decrease the bin size in the "frequency count" column and keep it at a value identical to the one I use in the histogram, the curve height is okay. However, now there's a new problem...the curve centre seems shifted at times compared to the histogram.
I am sending the file as per your suggestion.
Mily.
quote:
Originally posted by easwar
Hi Mily,
Please send the Origin project to tech support so we can take a look. Use the "Send File to Tech support" link on top right of this page, and refer to this post in your e-mail.
Easwar
OriginLab
|
| easwar |
Posted - 04/02/2012 : 10:49:24 AM
Hi Mily,
Please send the Origin project to tech support so we can take a look. Use the "Send File to Tech support" link on top right of this page, and refer to this post in your e-mail.
Easwar
OriginLab
|
|
|
