Any way to get contributions from control variables by channel/geo/time period?

[nataliaconnolly-alixpartners]

Hi All,

I am trying to use Meridian to understand the contributions of all the input variables (including control) to the output. I can't seem to figure out how to break down the baseline into its constituent components, including control variables.

To get the baseline contribution as a whole, I used

xr_ds = model_analysis.baseline_summary_metrics(confidence_level=0.9, aggregate_times=False, aggregate_geos=False)
baseline_df = xr_ds.sel(metric='mean', distribution='posterior').to_dataframe()
     

And now I am stuck - I don't see a way to break this down. Could someone please point me in the right direction?

Thank you!
Natalia