Calculating Error [June 22, 2004]
Posted: Tue Aug 24, 2004 8:34 pm
By Joel MacDougald on Tuesday, June 22, 2004 - 12:34 pm:
Good Afternoon Everyone;
I'm trying to figure out the best way for calculating the error in our HPLC preparations and was looking for resources. I have the basics for calculating percent error etc, but don't have anything for summing the errors. I have seen an equation that squared and added the percent error and then took the square root, but I can't find anything to explain this. Should I just do a straight forward addition? Any suggestions would be greatly appreciated.
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, June 23, 2004 - 02:37 am:
I guess it depends on what you are exactly looking for: comparing the errors of one analyst with an other, or variances of one method, ...
Please specify a bit more.
-------------------------------------------------------------------------------------------------------
By Joel MacDougald on Wednesday, June 23, 2004 - 06:05 am:
I would like to be able to compare variances in the methods, such as switching dilutions, weighing double, or even comparing one method to another. Note that I would like to be able to show the overall affect on error in the method, not just a comparison between two types of dilutions.
Thanks
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, June 23, 2004 - 12:31 pm:
Here is what FDA says
Robustness:
ICH defines robustness as a measure of the method's capability to remain unaffected by small, but deliberate variations in method
parameters. Robustness can be partly assured by good system suitability specifications. Thus, it is important to set tight, but realistic,system suitability specifications. Testing varying some or all conditions, e.g., age of columns, column type, column temperature, pH of buffer in mobile phase, reagents, is normally performed.
Recommendations:
Data obtained from studies for robustness, though not usually submitted, are recommended to be included as part of method validation.
-------------------------------------------------------------------------------------------------------
By Tom Mizukami on Thursday, June 24, 2004 - 12:32 am:
Hi Joel,
I think the NIST site is a good place to start, there are many papers on measurement uncertainty and error propagation. Standard error for class A volumetric glassware is available from any of the catalogues or the USP. Any undergrad text on analytical chem will have a chapter on this. Take a look here: http://physics.nist.gov/Pubs/guidelines/outline.html
Good luck.
-------------------------------------------------------------------------------------------------------
By Joel MacDougald on Monday, June 28, 2004 - 08:53 am:
Thanks to all who replied. Your help is greatly appreciated.
Good Afternoon Everyone;
I'm trying to figure out the best way for calculating the error in our HPLC preparations and was looking for resources. I have the basics for calculating percent error etc, but don't have anything for summing the errors. I have seen an equation that squared and added the percent error and then took the square root, but I can't find anything to explain this. Should I just do a straight forward addition? Any suggestions would be greatly appreciated.
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, June 23, 2004 - 02:37 am:
I guess it depends on what you are exactly looking for: comparing the errors of one analyst with an other, or variances of one method, ...
Please specify a bit more.
-------------------------------------------------------------------------------------------------------
By Joel MacDougald on Wednesday, June 23, 2004 - 06:05 am:
I would like to be able to compare variances in the methods, such as switching dilutions, weighing double, or even comparing one method to another. Note that I would like to be able to show the overall affect on error in the method, not just a comparison between two types of dilutions.
Thanks
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, June 23, 2004 - 12:31 pm:
Here is what FDA says
Robustness:
ICH defines robustness as a measure of the method's capability to remain unaffected by small, but deliberate variations in method
parameters. Robustness can be partly assured by good system suitability specifications. Thus, it is important to set tight, but realistic,system suitability specifications. Testing varying some or all conditions, e.g., age of columns, column type, column temperature, pH of buffer in mobile phase, reagents, is normally performed.
Recommendations:
Data obtained from studies for robustness, though not usually submitted, are recommended to be included as part of method validation.
-------------------------------------------------------------------------------------------------------
By Tom Mizukami on Thursday, June 24, 2004 - 12:32 am:
Hi Joel,
I think the NIST site is a good place to start, there are many papers on measurement uncertainty and error propagation. Standard error for class A volumetric glassware is available from any of the catalogues or the USP. Any undergrad text on analytical chem will have a chapter on this. Take a look here: http://physics.nist.gov/Pubs/guidelines/outline.html
Good luck.
-------------------------------------------------------------------------------------------------------
By Joel MacDougald on Monday, June 28, 2004 - 08:53 am:
Thanks to all who replied. Your help is greatly appreciated.