Hi Chris, In the ROOT fitting functions, two algorithms are currently implemented - normal Chisquare method (default) - likelihood fit (option "L") With the likelihood fit, all bins are taken into account. With the chisquare method, there are two cases: - default. Only bins with a non-null error are taken into account. In general, this means that all empty bins are ignored. However, you may have an empty bin, but have set its error to a non null value. Vice-versa, you may have a non-empty bin and set its error to 0. - option "W" is specified. In this case all bins are used and all have the same weight=1. Rene Brun Chris Theis wrote: > > Hi Rooters, > > I've been trying to figure out how the ROOT fitting routines actually > treat possible empty bins within a histogram. Unfortunately I've not been > able to find out if they are discarded or taken into account, respectively > if there is a switch to set the behavior. > > Probably one of you knows anything about this topic. Any hints would be > appreciated! > > Regards > Chris > > --------------------------------------------------------------------------- > Chris Theis > CERN/EP - European Organisation for Nuclear Research > 1211 Geneva 23, Switzerland > Phone: +41 22 767 6592 Office: 40-2B-24 > e-mail: Christian.Theis@cern.ch www: http://www.cern.ch/ctheis > --------------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:50:41 MET