Group Abstract Group Abstract

Message Boards Message Boards

Fitting with x and y errors?

I have a set of data points xi and yi with individual errors di and ei, and i make a linear fit through the data of the form y=ax+b. How do the errors di and ei translate to errors in a and b? So far i can only find erros in a and b based on the residuals.
POSTED BY: Mathijn Bernards
3 Replies
Sort By:

for instance if my error in my y data is huge i would expect the error in the parameters a and b to increase as well. will this happen in this case

That's the way I read the documentation. If you put in the errors on the data, and set the VarianceEstimatorFunction properly, then the calculated error in the parameters should depend only on the error of the data, not on the residuals to the fit.

Related discussions: http://community.wolfram.com/groups/-/m/t/875912 , http://mathematica.stackexchange.com/q/112568/9490

POSTED BY: Jason Biggs

Wouldn't this only help with the relative errors in the data? for instance if my error in my y data is huge i would expect the error in the parameters a and b to increase as well. will this happen in this case?
POSTED BY: Mathijn Bernards

This page may help: Fit Models with Measurement Errors

I think the relevant part for you may be

For measurement errors, you want standard errors to be computed only from the weights and so the variance estimate should be the constant 1.
POSTED BY: Jason Biggs
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Tag limit exceeded
Note: Only the first five people you tag will receive an email notification; the other tagged names will appear as links to their profiles.
Reply Preview
Attachments
Remove
or Discard