Calculation of decomps/ weekly contributions

[allisonwiener]

It looks like the weekly decomps by variable in pareto_alldecomp_matrix.csv output is calculated from the input data (with saturation/adstock applied) * coefficients.

1. Is that correct?
2. Does Robyn apply any transformations to the Xs or Y varaibles? Is it correct to do the decomp calculations without the transformation, but with the adstock/saturation applied? Thanks!