I have been working with the CausalImpact R-package to check whether new campaigns or features had a significant impact on a time-series. I would like to know if it is possible to quantify this impact (i.e. how many more units were sold due to this change) if the result had been a significant one.
This is exactly what is reported as 'Absolute effect' in the summary of a CausalImpact object, say,
(See the examples in the package documentation,