Sébastien Piérard

Toolkit: chained correlations

Quite often, a researcher has published the correlation between a variable A and another variable B, and another researcher has published the correlation between the variable B and a third variable C. In this case, it is not always necessary to do a new job to determine the correlation between A and C. There is a kind of transitivity of the correlations that allows to compute achievable bounds for the unknown correlation based on the two known correlations. You can freely use this tool to compute these bounds. This works both for Pearson and Spearman correlations.

Have fun! This tool is provided without any guarantee; please feel free to contact me if you find a bug.

correlation between A and B :
correlation between B and C :