[fluka-discuss]: Data merging statistics

http://www.fluka.org/web_archive/earchive/new-fluka-discuss/6181.html

From: Chenyen Lee <chenyenlee_at_gmail.com>
Date: Wed, 22 Jan 2014 09:59:08 +0800

Dear forum :
I would like to ask about a question about the data merging.
Performing a single run with 1e5 particles, I will get a statistical error
is +/- 99%.
But If I perform a continue run with 2 x 5e4 particles or 2 x 1e5
particles,
and then do the data merging,the statistical error is less than 1%.
Please advise me where I’m doing wrong ?

Thanks in advance

http://www.fluka.org/web_archive/earchive/new-fluka-discuss/att-6181/O18XS10.inp

 

http://www.fluka.org/web_archive/earchive/new-fluka-discuss/6194.html

From: <Andrea.Mairani_at_mi.infn.it>
Date: Sun, 26 Jan 2014 12:39:21 +0100

Dear Chenyen,
About statistical error you can read for example the last 4 slides of the
this presentation that explains how to interpret the results from a MC
calculation and it gives you some practical tips
(https://fluka-course.web.cern.ch/FLUKA-Course/Lectures_pdf/c04_StatSampl.pdf).
If you perform a single run by convention the statistical error is printed
as 99%.
Best Regards,
Andrea