ParetoNBDModel.expected_purchase_probability#

ParetoNBDModel.expected_purchase_probability(data=None, *, n_purchases=None, future_t=None)[source]#

Compute expected probability of n_purchases over future_t time periods.

Estimate probability of n_purchases over future_t time periods, given an individual customer’s current frequency, recency, and T.

Adapted from equation (16) in Bruce Hardie’s notes [1], and lifetimes package: CamDavidsonPilon/lifetimes

Parameters:

dataDataFrame

Optional dataframe containing the following columns:

customer_id: Unique customer identifier
frequency: Number of repeat purchases
recency: Time between the first and the last purchase
T: Time between the first purchase and the end of the observation period. Model assumptions require T >= recency
future_t: Optional column for future_t parametrization.
n_purchases: Optional column for n_purchases parametrization. Currently restricted to the same number for all customers.
All covariate columns specified when model was initialized.

If not provided, predictions will be ran with data used to fit model.

future_tarray_like

Number of time periods to predict expected purchases. Not required if data Dataframe contains a future_t column.

n_purchasesint

Number of purchases predicted. Not required if data Dataframe contains an n_purchases column.

future_tarray_like

Time periods over which the probability should be estimated. Not required if data Dataframe contains an n_purchases column.

References

[1]

Fader, Peter & G. S. Hardie, Bruce (2014). “Deriving the Conditional PMF of the Pareto/NBD Model.” https://www.brucehardie.com/notes/028/pareto_nbd_conditional_pmf.pdf