Calling Chi2Test on a few unweighted histograms I get the error message
In the code for TH1F (and in the reference documentation) it says
// igood=1'There is bin in 1st hist with low then 1 exp number of event
I’m not a native speaker, but I don’t think either is correct English. I think this is meant to say “There is a bin in the 1st histogram with less than 1 exp number of events”. But since I have no idea what “1 exp number” is, I also fail to understand what this is supposed to mean. The same goes for the other error messages.
While my not understanding may be due to my ignorance, the grammar problem should be fixed in any case.
the warning message comes when one of the histogram has zero entries. In that case the Chi2 test could be not very reliable. It is better either to combine the low statistics bin or performing un unbin tests on the data like the Kolmogorov-Smirnov test.
Thank you for reporting this, the documentation is now fixed in CVS.