A Unified Survey of Treatment Effect Heterogeneity Modelling and Uplift Modelling

A central question in many fields of scientific research is to determine how an outcome is affected by an action, i.e., to estimate the causal effect or treatment effect of an action. In recent years, in areas such as personalised healthcare, sociology, and online marketing, a need has emerged to estimate heterogeneous treatment effects with respect to individuals of different characteristics. To meet this need, two major approaches have been taken: treatment effect heterogeneity modelling and uplifting modelling. Researchers and practitioners in different communities have developed algorithms based on these approaches to estimate the heterogeneous treatment effects. In this article, we present a unified view of these two seemingly disconnected yet closely related approaches under the potential outcome framework. We provide a structured survey of existing methods following either of the two approaches, emphasising their inherent connections and using unified notation to facilitate comparisons. We also review the main applications of the surveyed methods in personalised marketing, personalised medicine, and sociology. Finally, we summarise and discuss the available software packages and source codes in terms of their coverage of different methods and applicability to different datasets, and we provide general guidelines for method selection.

Tax-Induced Inequalities in the Sharing Economy

The growth of sharing economy marketplaces like Airbnb has generated discussions on their socioeconomic impact and lack of regulation. As a result, most major cities in the United States have started to collect an “occupancy tax” for Airbnb bookings. In this study, we investigate the heterogeneous treatment effects of the occupancy tax policy on Airbnb listings, using a combination of a generalized causal forest methodology and a difference-in-differences framework. While we find that the introduction of the tax significantly reduces both listing revenues and sales, more importantly, these effects are disproportionately more pronounced for residential hosts with single shared-space (nontarget) listings versus commercial hosts with multiple properties or entire-space (target) listings. We further show that this unintended consequence is caused by customers’ discriminatory tax aversion against nontarget listings. We then leverage these empirical results by prescribing how hosts should optimally set prices in response to the occupancy tax and identify the discriminatory tax rates that would equalize the tax’s effect across nontarget and target listings. This paper was accepted by Victor Martínez-de-Albéniz, operations management.

Automated Interpretable Discovery of Heterogeneous Treatment Effectiveness: A Covid-19 Case Study

Testing multiple treatments for heterogeneous (varying) effectiveness with respect to many underlying risk factors requires many pairwise tests; we would like to instead automatically discover and visualize patient archetypes and predictors of treatment effectiveness using multitask machine learning. In this paper, we present a method to estimate these heterogeneous treatment effects with an interpretable hierarchical framework that uses additive models to visualize expected treatment benefits as a function of patient factors (identifying personalized treatment benefits) and concurrent treatments (identifying combinatorial treatment benefits). This method achieves state-of-the-art predictive power for Covid-19 in-hospital mortality and interpretable identification of heterogeneous treatment benefits. We first validate this method on the large public MIMIC-IV dataset of ICU patients to test recovery of heterogeneous treatment effects. Next we apply this method to a proprietary dataset of over 3000 patients hospitalized for Covid-19, and find evidence of heterogeneous treatment effectiveness predicted largely by indicators of inflammation and thrombosis risk: patients with few indicators of thrombosis risk benefit most from treatments against inflammation, while patients with few indicators of inflammation risk benefit most from treatments against thrombosis. This approach provides an automated methodology to discover heterogeneous and individualized effectiveness of treatments.

Medicaid, Earnings, and Heterogeneous Treatment Effects: Evidence from the Oregon Health Insurance Experiment

The Medicaid and labor supply empirical literature offers competing conclusions of zero effects and significant reductions in earnings. However, zero effects are only theoretically consistent with the earnings distribution’s extremes. Medicaid participants with positive pre-treatment labor supply should unequivocally decrease earnings. This paper clarifies the literature’s ambiguity by combining quantile regression with data from the Oregon Health Insurance Experiment. The distributional impacts imply that zero effects are not universally representative of Medicaid households. The annual earnings impact of Medicaid participation ranges between increases of $1400 to deceases of $3120 for single adults. Pre-existing mental illness or health constraints on work account for counterintuitive positive earnings impacts. By demonstrating that sample compositional differences determine whether Medicaid’s labor supply impact is zero or negative, this paper offers a reconciliation to the range of existing estimates in the empirical literature.

Deep Learning of Potential Outcomes

This review systematizes the emerging literature for causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction on how deep learning can be used to estimate/predict heterogeneous treatment effects and extend causal inference to settings where confounding is non-linear, time varying, or encoded in text, networks, and images. To maximize accessibility, we also introduce prerequisite concepts from causal inference and deep learning. The survey differs from other treatments of deep learning and causal inference in its sharp focus on observational causal estimation, its extended exposition of key algorithms, and its detailed tutorials for implementing, training, and selecting among deep estimators in Tensorflow 2 available at github.com/kochbj/Deep-Learning-for-Causal-Inference.

IDENTIFICATION OF REGRESSION MODELS WITH A MISCLASSIFIED AND ENDOGENOUS BINARY REGRESSOR

We study identification in nonparametric regression models with a misclassified and endogenous binary regressor when an instrument is correlated with misclassification error. We show that the regression function is nonparametrically identified if one binary instrument variable and one binary covariate satisfy the following conditions. The instrumental variable corrects endogeneity; the instrumental variable must be correlated with the unobserved true underlying binary variable, must be uncorrelated with the error term in the outcome equation, but is allowed to be correlated with the misclassification error. The covariate corrects misclassification; this variable can be one of the regressors in the outcome equation, must be correlated with the unobserved true underlying binary variable, and must be uncorrelated with the misclassification error. We also propose a mixture-based framework for modeling unobserved heterogeneous treatment effects with a misclassified and endogenous binary regressor and show that treatment effects can be identified if the true treatment effect is related to an observed regressor and another observable variable.