Hello.
I hope everything is OK despite the pandemic.
I wonder, since when are error bars being important on the weights of points the fitting curves? I just realized a fitting with error bars and non-error bars are different. Amazing
Fit with error bars
- MagicPlot
- Site Admin
- Posts: 67
- Joined: December 27th, 2018, 5:24 pm
- Location: St. Petersburg
- Contact:
Re: Fit with error bars
Hello,
All is OK, thank you, and I hope that you too.
Yes, the error data is used to calculate the weights of the points while fitting. The more error you have at a point, the less weight has it point -- it is a standard least-squares algorithm technique. The exact formulas are here (automatic links are disabled in forum to reduce spam): https://magicplot.com/wiki/fit_formulas
All is OK, thank you, and I hope that you too.
Yes, the error data is used to calculate the weights of the points while fitting. The more error you have at a point, the less weight has it point -- it is a standard least-squares algorithm technique. The exact formulas are here (automatic links are disabled in forum to reduce spam): https://magicplot.com/wiki/fit_formulas
Re: Fit with error bars
Cool, I did notice that before. Thanks.
Keep safe!
Keep safe!
Re: Fit with error bars
Seems like error in X is not considered. As I have seen in the program + https://magicplot.com/wiki/fit_formulas
Isn't it?
Isn't it?
- MagicPlot
- Site Admin
- Posts: 67
- Joined: December 27th, 2018, 5:24 pm
- Location: St. Petersburg
- Contact:
Re: Fit with error bars
Hello,
Yes, X errors are not taken into account by the current fit algorithm.
Yes, X errors are not taken into account by the current fit algorithm.