Covariance of two histograms

[Liuyao_Zhang]

@bellenot, Hi bellenot,
Yes, @moneta has helped me resolve how to propagate a uncertainty between two variables.
Actually, to get the uncertainty, a covariance between two variables is necessary.
I get a similar question like below. However, I need to recreate a new TH2. I wonder whether there is
a more simple method to get it.

Thanks a lot.
Liuyao

@awisecar

If you have distributions for two variables (each TH1’s) and want to find the correlation coefficient for these two variables, I’d imagine you’d create a TH2 with the appropriate number of bins for each axis, and then use the TH2::GetCorrelationFactor method:
https://root.cern.ch/doc/master/TH2_8cxx_source.html#l01025
which will return a double value.

Best,
Andrew