GammaGammaModelIndividual

class pymc_marketing.clv.models.gamma_gamma.GammaGammaModelIndividual(data, model_config=None, sampler_config=None)

Gamma-Gamma model

Estimate the average monetary value of customer transactions.

The model is conditioned on the individual transaction values per user, and is based on [1], [2].

TODO: Explain assumptions of model

Check GammaGammaModel for an equivalent model conditioned on the average transaction value of each user.

Parameters:

• data (pd.DataFrame) –

  Dataframe containing the following columns:

  • customer_id: Customer labels. The same value should be used for each observation

  coming from the same customer.

  • individual_transaction_value: Value of individual transactions.

• model_config (dict, optional) – Dictionary of model prior parameters. If not provided, the model will use default priors specified in the default_model_config class attribute.

• sampler_config (dict, optional) – Dictionary of sampler parameters. Defaults to None.

Examples

Gamma-Gamma model conditioned on individual customer spend

import pymc as pm
from pymc_marketing.clv import GammaGammaModelIndividual

model = GammaGammaModelIndividual(
    data=pd.DataFrame({
        "customer_id": [0, 0, 0, 1, 1, 2, ...],
        "individual_transaction_value": [5.3. 5.7, 6.9, 13.5, 0.3, 19.2 ...],
    }),
    model_config={
        "p_prior": {dist: 'HalfNorm', kwargs: {}},
        "q_prior": {dist: 'HalfStudentT', kwargs: {"nu": 4, "sigma": 10}},
        "v_prior": {dist: 'HalfCauchy', kwargs: {}},
    },
    sampler_config={
        "draws": 1000,
        "tune": 1000,
        "chains": 2,
        "cores": 2,
        "nuts_kwargs": {"target_accept": 0.95},
    },
)
model.build_model()
model.fit()
print(model.fit_summary())

# Predict spend of customers for which we know transaction history,
# conditioned on data. May include customers not included in fitting
expected_customer_spend = model.expected_customer_spend(
    customer_id=[0, 0, 0, 1, 1, 2, ...],
    individual_transaction_value=[5.3. 5.7, 6.9, 13.5, 0.3, 19.2 ...],
)
print(expected_customer_spend.mean("customer_id"))

# Predict spend of 10 new customers, conditioned on data
new_customer_spend = model.expected_new_customer_spend(n=10)
print(new_customer_spend.mean("new_customer_id"))

References

[1]

Fader, P. S., & Hardie, B. G. (2013). The Gamma-Gamma model of monetary value. February, 2, 1-9. http://www.brucehardie.com/notes/025/gamma_gamma.pdf

[2]

Peter S. Fader, Bruce G. S. Hardie, and Ka Lok Lee (2005), “RFM and CLV: Using iso-value curves for customer base analysis”, Journal of Marketing Research, 42 (November), 415-430. https://journals.sagepub.com/doi/pdf/10.1509/jmkr.2005.42.4.415

Methods

GammaGammaModelIndividual.__init__(data[, ...])	Initializes model configuration and sampler configuration for the model
GammaGammaModelIndividual.build_model()	Creates an instance of pm.Model based on provided data and model_config, and attaches it to self.
GammaGammaModelIndividual.distribution_customer_spend(...)	Return distribution of transaction value per customer
GammaGammaModelIndividual.distribution_new_customer_spend([...])	Posterior distribution of transaction value for new customers
GammaGammaModelIndividual.expected_customer_lifetime_value(...)	Return expected customer lifetime value.
GammaGammaModelIndividual.expected_customer_spend(...)	Return expected transaction value per customer
GammaGammaModelIndividual.expected_new_customer_spend()	Expected transaction value for a new customer
GammaGammaModelIndividual.fit([fit_method])	Infer model posterior
GammaGammaModelIndividual.fit_summary(**kwargs)
GammaGammaModelIndividual.get_params([deep])	Get all the model parameters needed to instantiate a copy of the model, not including training data.
GammaGammaModelIndividual.load(fname)	Creates a ModelBuilder instance from a file, Loads inference data for the model.
GammaGammaModelIndividual.predict(X_pred[, ...])	Uses model to predict on unseen data and return point prediction of all the samples.
GammaGammaModelIndividual.predict_posterior(X_pred)	Generate posterior predictive samples on unseen data.
GammaGammaModelIndividual.predict_proba(X_pred)	Alias for predict_posterior, for consistency with scikit-learn probabilistic estimators.
GammaGammaModelIndividual.sample_posterior_predictive(X_pred)	Sample from the model's posterior predictive distribution.
GammaGammaModelIndividual.sample_prior_predictive(X_pred)	Sample from the model's prior predictive distribution.
GammaGammaModelIndividual.save(fname)	Save the model's inference data to a file.
GammaGammaModelIndividual.set_idata_attrs([idata])	Set attributes on an InferenceData object.
GammaGammaModelIndividual.set_params(**params)	Set all the model parameters needed to instantiate the model, not including training data.
GammaGammaModelIndividual.thin_fit_result(...)	Return a copy of the model with a thinned fit result.

Attributes

X
default_model_config	Returns a class default config dict for model builder if no model_config is provided on class initialization Useful for understanding structure of required model_config to allow its customization by users .
default_sampler_config	Returns a class default sampler dict for model builder if no sampler_config is provided on class initialization Useful for understanding structure of required sampler_config to allow its customization by users .
fit_result
id	Generate a unique hash value for the model.
output_var	Returns the name of the output variable of the model.
version
y