How should I solve the case where the two experimental values are correlated,in PYMC3?

Suppose I have three experimental values ABC, A and B are independent experimental values, but A and C come from the same data. At this time, I need to use the experimental values of ABC to fit parameter X. So what do I have to do with the correlation between A and B to make it more statistically significant?In the more complicated case, only part of A and C are derived from the same data, and the rest are derived from unrelated data. In other words, in addition to having the same data, the sources of A and C are also averaged with other experimental results, so the correlation between A and C is not 100%. In this case, how do I adjust some of the inputs in PYMC3 to fit?