BaseMMM.plot_grouped_contribution_breakdown_over_time

BaseMMM.plot_grouped_contribution_breakdown_over_time(stack_groups=None, original_scale=False, area_kwargs=None, **plt_kwargs)

Plot a time series area chart for all channel contributions.

Since a chart like this can become quite crowded if you have many channels or control variables, you can group certain variables together using the stack_groups keyword.

Parameters:

stack_groupsdict of {str: list of str}, optional

Specifies which variables to group together. Example: passing

{

“Baseline”: [“intercept”], “Offline”: [“TV”, “Radio”], “Online”: [“Banners”]

}

results in a chart with three colors, one for Baseline, one for Online, and one for Offline. If stack_groups is None, the chart would have four colors since TV and Radio would be separated.

Note: If you only pass {“Baseline”: “intercept”, “Online”: [“Banners”]}, you will not see the TV and Radio channels in the chart.

original_scalebool, by default False

If True, the contributions are plotted in the original scale of the target.

Returns:

plt.Figure

Matplotlib figure with the plot.