Saved in:
Bibliographic Details
Main Authors: Verhelst, Théo, Bontempi, Gianluca
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.08805
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909952044433408
author Verhelst, Théo
Bontempi, Gianluca
author_facet Verhelst, Théo
Bontempi, Gianluca
contents Uplift modeling estimates the causal effect of an intervention as the difference between potential outcomes under treatment and control, whereas counterfactual identification aims to recover the joint distribution of these potential outcomes (e.g., "Would this customer still have churned had we given them a marketing offer?"). This joint counterfactual distribution provides richer information than the uplift but is harder to estimate. However, the two approaches are synergistic: uplift models can be leveraged for counterfactual estimation. We propose a counterfactual estimator that fits a bivariate beta distribution to predicted uplift scores, yielding posterior distributions over counterfactual outcomes. Our approach requires no causal assumptions beyond those of uplift modeling. Simulations show the efficacy of the approach, which can be applied, for example, to the problem of customer churn in telecom, where it reveals insights unavailable to standard ML or uplift models alone.
format Preprint
id arxiv_https___arxiv_org_abs_2512_08805
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Identifying counterfactual probabilities using bivariate distributions and uplift modeling
Verhelst, Théo
Bontempi, Gianluca
Machine Learning
G.3, I.2.0
Uplift modeling estimates the causal effect of an intervention as the difference between potential outcomes under treatment and control, whereas counterfactual identification aims to recover the joint distribution of these potential outcomes (e.g., "Would this customer still have churned had we given them a marketing offer?"). This joint counterfactual distribution provides richer information than the uplift but is harder to estimate. However, the two approaches are synergistic: uplift models can be leveraged for counterfactual estimation. We propose a counterfactual estimator that fits a bivariate beta distribution to predicted uplift scores, yielding posterior distributions over counterfactual outcomes. Our approach requires no causal assumptions beyond those of uplift modeling. Simulations show the efficacy of the approach, which can be applied, for example, to the problem of customer churn in telecom, where it reveals insights unavailable to standard ML or uplift models alone.
title Identifying counterfactual probabilities using bivariate distributions and uplift modeling
topic Machine Learning
G.3, I.2.0
url https://arxiv.org/abs/2512.08805