MVITS

class pymc_marketing.customer_choice.mv_its.MVITS(existing_sales, saturated_market=True, model_config=None, sampler_config=None)

Multivariate Interrupted Time Series class.

Class to perform a multivariate interrupted time series analysis with the specific intent of determining where the sales of a new product came from.

Parameters:

existing_saleslist of str

The names of the existing products.

saturated_marketbool, optional

Whether the market is saturated or not. If True, the sum of the beta’s will be 1. Else, the sum of the beta’s will be less than 1 with the remaining sales attributed to the new product.

model_configdict, optional

The model configuration. If None, the default model configuration will be used.

sampler_configdict, optional

The sampler configuration. If None, the default sampler configuration will be used.

Methods

MVITS.__init__(existing_sales[, ...])	Initialize model configuration and sampler configuration for the model.
MVITS.attrs_to_init_kwargs(attrs)	Convert the attributes of the InferenceData object to the __init__ kwargs.
MVITS.build_from_idata(idata)	Build model from the InferenceData object.
MVITS.build_model(X, y, **kwargs)	Build a PyMC model for a multivariate interrupted time series analysis.
MVITS.calculate_counterfactual([random_seed])	Calculate the counterfactual scenario of never releasing the new product.
MVITS.causal_impact([variable])	Calculate the causal impact of the new product on the existing products.
MVITS.create_fit_data(X, y)	Create the fit_data group based on the input data.
MVITS.create_idata_attrs()	Create the attributes for the InferenceData object.
MVITS.fit(X[, y, progressbar, random_seed])	Fit a model using the data passed as a parameter.
MVITS.graphviz(**kwargs)	Get the graphviz representation of the model.
MVITS.inform_default_prior(data)	Inform the default prior based on the data.
MVITS.load(fname)	Create a ModelBuilder instance from a file.
MVITS.load_from_idata(idata)	Create a ModelBuilder instance from an InferenceData object.
MVITS.plot_causal_impact_market_share([...])	Plot the inferred causal impact of the new product on the existing products.
MVITS.plot_causal_impact_sales([variable, ax])	Plot causal impact of sales.
MVITS.plot_counterfactual([variable, ...])	Plot counterfactual scenario.
MVITS.plot_data([plot_total_sales, ax])	Plot the observed data.
MVITS.plot_fit([variable, plot_total_sales, ax])	Plot the model fit (posterior predictive) of the existing products.
MVITS.post_sample_model_transformation()	Perform transformation on the model after sampling.
MVITS.predict([X, extend_idata])	Use a model to predict on unseen data and return point prediction of all the samples.
MVITS.predict_posterior([X, extend_idata, ...])	Generate posterior predictive samples on unseen data.
MVITS.predict_proba([X, extend_idata, combined])	Alias for predict_posterior, for consistency with scikit-learn probabilistic estimators.
MVITS.sample(X, y[, random_seed, ...])	Sample all the things.
MVITS.sample_posterior_predictive([X, ...])	Sample from the model's posterior predictive distribution.
MVITS.sample_prior_predictive([X, y, ...])	Sample from the model's prior predictive distribution.
MVITS.save(fname)	Save the model's inference data to a file.
MVITS.set_idata_attrs([idata])	Set attributes on an InferenceData object.

Attributes

X
default_model_config	Default model configuration.
default_sampler_config	Default sampler configuration.
fit_result	Get the posterior fit_result.
id	Generate a unique hash value for the model.
output_var	The output variable of the model.
posterior
posterior_predictive
predictions
prior
prior_predictive
version
y