Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
2,187
result(s) for
"Counterfactuals"
Sort by:
HUMAN DECISIONS AND MACHINE PREDICTIONS
Can machine learning improve human decision making? Bail decisions provide a good test case. Millions of times each year, judges make jail-or-release decisions that hinge on a prediction of what a defendant would do if released. The concreteness of the prediction task combined with the volume of data available makes this a promising machine-learning application. Yet comparing the algorithm to judges proves complicated. First, the available data are generated by prior judge decisions. We only observe crime outcomes for released defendants, not for those judges detained. This makes it hard to evaluate counterfactual decision rules based on algorithmic predictions. Second, judges may have a broader set of preferences than the variable the algorithm predicts; for instance, judges may care specifically about violent crimes or about racial inequities. We deal with these problems using different econometric strategies, such as quasi-random assignment of cases to judges. Even accounting for these concerns, our results suggest potentially large welfare gains: one policy simulation shows crime reductions up to 24.7% with no change in jailing rates, or jailing rate reductions up to 41.9% with no increase in crime rates. Moreover, all categories of crime, including violent crimes, show reductions; these gains can be achieved while simultaneously reducing racial disparities. These results suggest that while machine learning can be valuable, realizing this value requires integrating these tools into an economic framework: being clear about the link between predictions and decisions; specifying the scope of payoff functions; and constructing unbiased decision counterfactuals.
Journal Article
Eye-Tracking Causality
How do people make causal judgments? What role, if any, does counterfactual simulation play? Counterfactual theories of causal judgments predict that people compare what actually happened with what would have happened if the candidate cause had been absent. Process theories predict that people focus only on what actually happened, to assess the mechanism linking candidate cause and outcome. We tracked participants' eye movements while they judged whether one billiard ball caused another one to go through a gate or prevented it from going through. Both participants' looking patterns and their judgments demonstrated that counterfactual simulation played a critical role. Participants simulated where the target ball would have gone if the candidate cause had been removed from the scene. The more certain participants were that the outcome would have been different, the stronger the causal judgments. These results provide the first direct evidence for spontaneous counterfactual simulation in an important domain of high-level cognition.
Journal Article
Improving the Interpretation of Fixed Effects Regression Results
Fixed effects estimators are frequently used to limit selection bias. For example, it is well known that with panel data, fixed effects models eliminate time-invariant confounding, estimating an independent variable’s effect using only within-unit variation. When researchers interpret the results of fixed effects models, they should therefore consider hypothetical changes in the independent variable (counterfactuals) that could plausibly occur within units to avoid overstating the substantive importance of the variable’s effect. In this article, we replicate several recent studies which used fixed effects estimators to show how descriptions of the substantive significance of results can be improved by precisely characterizing the variation being studied and presenting plausible counterfactuals. We provide a checklist for the interpretation of fixed effects regression results to help avoid these interpretative pitfalls.
Journal Article
INFERENCE ON COUNTERFACTUAL DISTRIBUTIONS
Counterfactual distributions are important ingredients for policy analysis and decomposition analysis in empirical economics. In this article, we develop modeling and inference tools for counterfactual distributions based on regression methods. The counterfactual scenarios that we consider consist of ceteris paribus changes in either the distribution of covariates related to the outcome of interest or the conditional distribution of the outcome given covariates. For either of these scenarios, we derive joint functional central limit theorems and bootstrap validity results for regression-based estimators of the status quo and counterfactual outcome distributions. These results allow us to construct simultaneous confidence sets for function-valued effects of the counterfactual changes, including the effects on the entire distribution and quantile functions of the outcome as well as on related functionals. These confidence sets can be used to test functional hypotheses such as no-effect, positive effect, or stochastic dominance. Our theory applies to general counterfactual changes and covers the main regression methods including classical, quantile, duration, and distribution regressions. We illustrate the results with an empirical application to wage decompositions using data for the United States. As a part of developing the main results, we introduce distribution regression as a comprehensive and flexible tool for modeling and estimating the entire conditional distribution. We show that distribution regression encompasses the Cox duration regression and represents a useful alternative to quantile regression. We establish functional central limit theorems and bootstrap validity results for the empirical distribution regression process and various related functionals.
Journal Article
Phantom Counterfactuals
Researchers often seek to identify the effects of a treatment on a sequence of behaviors, such as whether citizens register to vote and whether they then cast ballots. I show that average treatment effects (ATEs) are only identified until the first behavior (registering to vote) that affects the set of possible subsequent actions (voting). When one action changes the set of possible subsequent actions, it creates ‘phantom counterfactuals,’or undefined potential outcomes, which render ATEs unidentified. I show that applied theory allows researchers to diagnose phantom counterfactuals, which helps to recognize unidentified ATEs and focus instead on other estimands that are identified. I illustrate this approach using a stylized model of crime reporting, showing how different theories generate different sets of identified estimands while holding constant an experimental design. I thereby establish the necessity of applied theory for causal identification in empirical research with sequential behavioral outcomes.
Journal Article
Inferences from the negation of counterfactual and semifactual conditionals
Our goal was to study how people understand the negation of counterfactuals (such as “Antonio
denied
/
said that it is false
that if Messi had played, then Barcelona would have won”) and semifactuals (such as “Antonio
denied that
even if Messi had played, Barcelona would have won”). Previous studies have shown that participants negated basic conditionals using small-scope interpretations by endorsing a new conditional with the negated consequent, but also by making large-scope interpretations, endorsing a conjunction with the negated consequent. Three experiments showed that when participants were asked whether the negation of a counterfactual (Experiments
1
and
2
) or semifactual (Experiment
3
) conditional was followed by a new conditional, they made a small-scope interpretation, endorsing the same conditional with the negated consequent (e.g., “if/even if Messi had played, Barcelona would
not
have won”). However, they also accepted the conditional with the negated antecedent for semifactuals (e.g., “even if Messi had
not
played, Barcelona would have won”). When participants were asked whether the negation of a counterfactual or semifactual conditional is followed by a conjunction, they endorsed the conjunction with both the negated antecedent and the consequent (e.g., “Messi did
not
play and Barcelona did
not
win”), but again they accepted the conjunction with the negated antecedent only for semifactuals (e.g., “Messi did
not
play and Barcelona did win”). These results have implications for the main theories of reasoning.
Journal Article
FOUNDATIONS OF STRUCTURAL CAUSAL MODELS WITH CYCLES AND LATENT VARIABLES
Structural causal models (SCMs), also known as (nonparametric) structural equation models (SEMs), are widely used for causal modeling purposes. In particular, acyclic SCMs, also known as recursive SEMs, form a well-studied subclass of SCMs that generalize causal Bayesian networks to allow for latent confounders. In this paper, we investigate SCMs in a more general setting, allowing for the presence of both latent confounders and cycles. We show that in the presence of cycles, many of the convenient properties of acyclic SCMs do not hold in general: they do not always have a solution; they do not always induce unique observational, interventional and counterfactual distributions; a marginalization does not always exist, and if it exists the marginal model does not always respect the latent projection; they do not always satisfy a Markov property; and their graphs are not always consistent with their causal semantics. We prove that for SCMs in general each of these properties does hold under certain solvability conditions. Our work generalizes results for SCMs with cycles that were only known for certain special cases so far. We introduce the class of simple SCMs that extends the class of acyclic SCMs to the cyclic setting, while preserving many of the convenient properties of acyclic SCMs. With this paper, we aim to provide the foundations for a general theory of statistical causal modeling with SCMs.
Journal Article