Evaluating and Correcting Performative Effects of Decision Support Systems via Causal Domain Shift

TL;DR

This paper addresses the performative effects of decision support systems by modeling their deployment as causal domain shift, proposing methods to assess and correct biases introduced by their influence in high-stakes settings.

Contribution

It introduces a causal domain shift framework for evaluating and correcting performative effects of decision support systems, with novel identification results for conditional expectations.

Findings

Repeated regression effectively estimates causal effects under bias.

The approach handles sample selection bias and selective labeling.

Empirical results demonstrate practical utility in real-world scenarios.

Abstract

When predicting a target variable $Y$ from features $X$, the prediction $\hat{Y}$ can be performative: an agent might act on this prediction, affecting the value of $Y$ that we eventually observe. Performative predictions are deliberately prevalent in algorithmic decision support, where a Decision Support System (DSS) provides a prediction for an agent to affect the value of the target variable. When deploying a DSS in high-stakes settings (e.g. healthcare, law, predictive policing, or child welfare screening) it is imperative to carefully assess the performative effects of the DSS. In the case that the DSS serves as an alarm for a predicted negative outcome, naive retraining of the prediction model is bound to result in a model that underestimates the risk, due to effective workings of the previous model. In this work, we propose to model the deployment of a DSS as causal domain shift and provide novel cross-domain identification results for the conditional expectation $E[Y | X]$, allowing for pre- and post-hoc assessment of the deployment of the DSS, and for retraining of a model that assesses the risk under a baseline policy where the DSS is not deployed. Using a running example, we empirically show that a repeated regression procedure provides a practical framework for estimating these quantities, even when the data is affected by sample selection bias and selective labelling, offering for a practical, unified solution for multiple forms of target variable bias.