MMMSummaryFactory.saturation_scatterplot

MMMSummaryFactory.saturation_scatterplot(original_scale=True, apply_cost_per_unit=True, dims=None, output_format=None)

Create saturation scatterplot data as a long-form DataFrame.

Parameters:

original_scalebool, default True

If True, plot contributions in original (un-scaled) units.

apply_cost_per_unitbool, default True

If True and cost-per-unit data is available, x is spend. Otherwise x is raw channel data.

dimsdict, optional

Dimension filters, e.g. {"channel": ["tv"]}.

output_format{“pandas”, “polars”}, optional

Output DataFrame format (default: uses factory default)

Returns:

pd.DataFrame or pl.DataFrame

Long-format DataFrame with columns:

• date: Time index

• channel: Channel name

• x: Spend or channel data value

• y: Mean posterior channel contribution

• <custom_dims>: One column per custom dimension when present