6
$\begingroup$

In principle a simple question: What is the pull distribution?

(All I could find out is that it is the error-weighted distribution of estimators around the true value.)

I'd be interested in the precise mathematical definition, how, why, and when to use it, what is it expected to look like, and if both estimator values and "true" value have errors associated to them, do these add up in quadrature.

PS: Feel free to add / change tags, I didn't find any good ones.

$\endgroup$
  • 2
    $\begingroup$ Can you indicate where you encountered this term? $\endgroup$ – Nick Sabbe May 26 '11 at 13:42
  • 1
    $\begingroup$ It may be particle physicists' slang. E.g. cmsdoc.cern.ch/documents/99/note99_029.pdf, root.cern.ch/phpBB3/viewtopic.php?t=7394. I'm using it myself in the sense stated above, but I'd like to know if that's correct and if there's more to it. $\endgroup$ – fuenfundachtzig May 26 '11 at 14:06
  • 1
    $\begingroup$ How about mailing the authors of the paper you found the term in? $\endgroup$ – Stephan Kolassa May 27 '11 at 19:31
  • 2
    $\begingroup$ Well, you know, it's one of those terms everybody uses so very casually that you'd feel ashamed to ask what it is. But apparently, the question is legitimate :) $\endgroup$ – fuenfundachtzig May 28 '11 at 11:02
3
$\begingroup$

I think this CDF public analysis note can be a good answer.

or the ps file from CDF.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.