Tools for Customer Lifetime Value Estimation
Coerce to clv.data object
Coerce to a Data Frame
Coerce to a Data Table
BG/BB models - Work In Progress
BG/NBD: Conditional Expected Transactions
BG/NBD: Unconditional Expectation
BG/NBD: Log-Likelihood functions
BG/NBD: Probability of Being Alive
BG/NBD: Probability Mass Function (PMF)
BG/NBD models
Result of fitting the BG/NBD model without covariates
Result of fitting the BG/NBD model with static covariates
Bootstrapping: Fit a model again on sampled data and apply method
Transactional data to fit CLV models
Transactional and dynamic covariates data to fit CLV models
Transactional and static covariates data to fit CLV models
Fitted model without covariates
Fitted Spending Model
Fitted Transaction Model without covariates
Fitted CLV Model with Dynamic covariates
Fitted Transaction Model with Static covariates
Result of fitting the Gamma-Gamma model
Result of fitting the GGompertz/NBD model without covariates
Result of fitting the GGompertz/NBD model with static covariates
CLV Model providing model related functionalities
CLV Model functionality for BG/NBD without covariates
CLV Model functionality for BG/NBD with static covariates
CLV Model functionality for the Gamma-Gamma spending model
CLV Model functionality for GGompertz/NBD without covariates
CLV Model functionality for GGompertz/NBD with static covariates
CLV Model without support for life-trans correlation
CLV Model functionality for PNBD with dynamic covariates
CLV Model functionality for Pareto/NBD without covariates
CLV Model functionality for Pareto/NBD with static covariates
CLV Model providing life-trans correlation related functionalities
Result of fitting the Pareto/NBD model without covariates
Result of fitting the Pareto/NBD model with dynamic covariates
Result of fitting the Pareto/NBD model with static covariates
Time Unit defining conceptual periods
Date based time-units
POSIXct based time-units
Time unit representing a single Day
Time unit representing a single hour
Time unit representing a single Week
Time unit representing a single Year
Create an object for transactional data required to estimate CLV
Customer Lifetime Value Tools
Extract Unconditional Expectation
Gamma-Gamma: Log-Likelihood Function
Gamma/Gamma Spending model
GGompertz/NBD: Conditional Expected Transactions
GGompertz/NBD: Unconditional Expectation
GGompertz/NBD: Log-Likelihood functions
GGompertz/NBD: Probability of Being Alive
GGompertz/NBD: Probability Mass Function (PMF)
Gamma-Gompertz/NBD model
Calculate hessian for a fitted model
Formula Interface for Latent Attrition Models
Likelihood Ratio Test of Nested Models
New customer prediction data
Number of observations
Number of observations
Plot Diagnostics for the Transaction data in a clv.data Object
Plot expected and actual mean spending per transaction
Plot Diagnostics for a Fitted Transaction Model
Probability Mass Function
Pareto/NBD: Conditional Expected Transactions
Pareto/NBD: Discounted Expected Residual Transactions
Pareto/NBD: Unconditional Expectation
Pareto/NBD: Log-Likelihood functions
Pareto/NBD: Probability of Being Alive
Pareto/NBD: Probability Mass Function (PMF)
Pareto/NBD models
Infer customers' spending
Predict CLV from a fitted transaction model
Add Dynamic Covariates to a CLV data object
Add Static Covariates to a CLV data object
Formula Interface for Spending Models
Subsetting clv.data
Summarizing a CLV data object
Summarizing a fitted CLV model
Summarizing a CLV time object
Calculate Variance-Covariance Matrix for CLV Models fitted with Maximu...
GSL Hypergeometric 2F0 for equal length vectors
GSL Hypergeometric 2F1 for equal length vectors
A set of state-of-the-art probabilistic modeling approaches to derive estimates of individual customer lifetime values (CLV). Commonly, probabilistic approaches focus on modelling 3 processes, i.e. individuals' attrition, transaction, and spending process. Latent customer attrition models, which are also known as "buy-'til-you-die models", model the attrition as well as the transaction process. They are used to make inferences and predictions about transactional patterns of individual customers such as their future purchase behavior. Moreover, these models have also been used to predict individuals’ long-term engagement in activities such as playing an online game or posting to a social media platform. The spending process is usually modelled by a separate probabilistic model. Combining these results yields in lifetime values estimates for individual customers. This package includes fast and accurate implementations of various probabilistic models for non-contractual settings (e.g., grocery purchases or hotel visits). All implementations support time-invariant covariates, which can be used to control for e.g., socio-demographics. If such an extension has been proposed in literature, we further provide the possibility to control for time-varying covariates to control for e.g., seasonal patterns. Currently, the package includes the following latent attrition models to model individuals' attrition and transaction process: [1] Pareto/NBD model (Pareto/Negative-Binomial-Distribution), [2] the Extended Pareto/NBD model (Pareto/Negative-Binomial-Distribution with time-varying covariates), [3] the BG/NBD model (Beta-Gamma/Negative-Binomial-Distribution) and the [4] GGom/NBD (Gamma-Gompertz/Negative-Binomial-Distribution). Further, we provide an implementation of the Gamma/Gamma model to model the spending process of individuals.
Useful links