TimeSliceCrossValidator

class pymc_marketing.mmm.time_slice_cross_validation.TimeSliceCrossValidator(n_init, forecast_horizon, date_column, step_size=1, sampler_config=None)

Time-Slice Cross Validator for Media Mix Models (MMM).

Provides scikit-learn-style API:

• split(X, y): yields train/test indices

• get_n_splits(): returns number of splits

• run(): executes full CV loop and returns fitted models and predictions

Methods

TimeSliceCrossValidator.__init__(n_init, ...)
TimeSliceCrossValidator.get_n_splits(X[, y])	Return number of possible rolling splits.
TimeSliceCrossValidator.run(X, y[, ...])	Run the complete time-slice CV loop.
TimeSliceCrossValidator.split(X, y)	Yield (train_idx, test_idx) pairs for each time-slice split.

Attributes

plot	Use the MMMPlotSuite to plot the results.