RE: Ask: the error of the mean value of a weighted histogram

From: Amnon Harel <amnon.harel_at_cern.ch>
Date: Thu, 23 Feb 2012 09:18:39 +0000


https://savannah.cern.ch/bugs/?19497



From: Lei Zhang
Sent: 23 February 2012 10:05
To: Amnon Harel
Cc: roottalk_at_root.cern.ch
Subject: Re: [ROOT] Ask: the error of the mean value of a weighted histogram

Dear Amnon,

Thanks for you reply.

Since I am not familiar with the webpage of the root bugs, I browsed over the bug list, but didn't find the bug#19494.

Could you please send me the link?
Sorry for this bother.

Cheers

Lei

On 2/23/2012 9:21, Amnon Harel wrote:
> Dear Lei,
>
> Perhaps the discussion of ROOT bug #19497
> answers your question.
>
> cheers,
> Amnon
> ________________________________________
> From: owner-roottalk_at_root.cern.ch [owner-roottalk_at_root.cern.ch] on behalf of Lei Zhang [lei.zhang_at_cern.ch]
> Sent: 23 February 2012 09:11
> To: roottalk_at_root.cern.ch
> Subject: [ROOT] Ask: the error of the mean value of a weighted histogram
>
> Dear ROOTors,
>
> According to my study, I need to get the error of the mean value of a weighted histogram.
>
> I have checked the code for calculating this error:
>
> (from the root web) // mean error = RMS / sqrt( Neff )
>
> The Neff is: (sum of Wt)^2 / sum of Wt^2
>
> // s[1] sum of squares of weights, s[0] sum of weights
> return (s[1] ? s[0]*s[0]/s[1] : TMath<http://root.cern.ch/root/html/TMath.html>::Abs<http://root.cern.ch/root/html/TMath.html#TMath:Abs>(s[0]) );
>
> Since I am not an expert on statistics, may I ask how this definition come? Where can I find some reference for this formula?
>
> Thank you in advance
>
> Best regards
>
> Lei
>
>
Received on Thu Feb 23 2012 - 10:18:46 CET

This archive was generated by hypermail 2.2.0 : Thu Feb 23 2012 - 23:50:01 CET