arXiv subjects

Charles F. Manski

Publications and source records attributed to Charles F. Manski.

At least 19 recordsLinked to original sources

arXiv2026

Credible Discourse on Climate Policy: Beyond Dueling Certitudes

Whatever the policy question under consideration, reasoned and realistic evaluation requires credible policy analysis under uncertainty. This holds particularly to the study of climate policy in the United States, where science and ideology have become increasingly entangled. Welfare economics provides a transparent formal framework to describe and evaluate the subtle tradeoffs that must be addressed when comparing alternative climate policies. The largely qualitative discussions of climate policy in public-facing reports suffer from failure to use welfare economics to assess tradeoffs. I use 2025 reports by the National Academies and by the US Department of Energy as case studies.

econ.GN↗

arXiv2026

Systemic Methodological Dysfunction in Statistical Research for Clinical Decisions

I critique a set of entrenched methodological conventions that collectively create systemic dysfunction in statistical research for clinical decisions. These include: (1) the prevalent use of hypothesis tests to compare treatments, (2) remoteness from patient care of the methods used to evaluate the accuracy of predictions of patient outcomes, (3) poor practice of meta-analysis to combine findings across studies, and (4) widespread research with incredible certitude. It appears that the dysfunction is held in place by three factors: (i) rudimentary instruction in statistical methodology received by medical students and residents, (ii) reliance of clinical researchers on consulting biostatisticians, wo act as statistical gatekeepers in evaluation of grant proposals and paper submissions, and (iii) institutional practices of research funding agencies, medical journals, and governmental bodies that regulate medical treatment. I conjecture that systemic changes are necessary to break the existing impasse, moving statistical research to a better equilibrium.

econ.EM↗

arXiv2026

Coping with Inductive Risk When Theories are Underdetermined: Decision Making with Partial Identification

Controversy about the significance of underdetermination of theories persists in the philosophy and conduct of science. The issue has practical import when research is used to inform decision making, because scientific uncertainty yields inductive risk. Seeking to enhance communication between philosophers and researchers who study public policy, this paper describes econometric analysis of partial identification and its use in welfare-economic policy analysis. Study of partial identification finds underdetermination and inductive risk to be highly consequential for credible prediction of important societal outcomes and, hence, for credible public decision making. It provides mathematical tools to characterize a broad class of scientific uncertainties that arise when available data and well-supported assumptions are combined to predict population outcomes. Combining study of partial identification with criteria for reasonable decision making under uncertainty yields coherent approaches to make policy choices without accepting one among multiple empirically underdetermined theories. The paper argues that study of partial identification warrants attention in philosophical discourse on underdetermination and inductive risk.

econ.EM↗

arXiv2025

Regret in Treatment Choice when Welfare Varies with an Uncertain Event: The Prediction-Threshold Problem

We study maximum regret (MR) of binary treatment choice in a population with observed covariates x, when welfare varies with an uncertain binary event. We consider decision making with plug-in probabilistic predictions of the event and pre-specified decision thresholds, which we term the prediction-threshold problem. The optimal treatment for persons with covariate value x is B if the conditional probability P(y=1|x) of a binary outcome y exceeds a particular x-specific threshold and is A otherwise. This structure is common in medical decision making and other contexts. Plug-in prediction uses data to estimate P(y|x) and acts as if the estimate is accurate. However, plug-in prediction is often performed with misspecified prediction models and conventional x-invariant thresholds. We use a combination of algebraic and computational analysis of limit and finite-sample MR to demonstrate how MR depends on the prediction model, the state space, and the thresholds used to choose treatments.

econ.EM↗

arXiv2025

Prediction with Differential Covariate Classification: Illustrated by Covariate Classification in Medical Risk Assessment

A common practice in evidence-based decision-making uses estimates of conditional probabilities P(y|x) obtained from research studies to predict outcomes y on the basis of observed covariates x. Given this information, decisions are then based on the predicted outcomes. Researchers commonly assume that the predictors used in the generation of the evidence are the same as those used in applying the evidence: i.e., the meaning of x in the two circumstances is the same. This may not be the case in real-world settings. Across a wide range of settings, ranging from clinical practice to education policy, demographic attributes (e.g., age, race, ethnicity) are often classified differently in research studies than in decision settings. This paper studies identification in such settings. We propose a formal framework for prediction with what we term differential covariate classification (DCC). Using this framework, we analyze partial identification of probabilistic predictions and assess how various assumptions influence the identification regions. We apply the findings to a range of settings, focusing mainly on differential classification of individuals' race and ethnicity in clinical medicine. We find that bounds on P(y|x) can be wide, and the information needed to narrow them available only in special cases. These findings highlight an important problem in using evidence in decision making, a problem that has not yet been fully appreciated in debates on classification in public policy and medicine.

econ.EM↗

arXiv2025

What is the general Welfare? Welfare Economic Perspectives on Equity

Researchers do not know what the framers of the United States Constitution intended when they wrote of the general Welfare. Nevertheless, economists can conjecture by specifying social welfare functions that aim to express the preferences of the population. Economists have often simplified analysis of public policy by assuming that individuals have homogeneous, consequentialist, and self-centered preferences. In reality, individuals may hold heterogeneous private and distributional preferences. To enhance policy analysis, economists should specify social welfare functions that express the richness and variety of actual personal preferences. The possibilities are vast. I focus on preferences for equity. There has been much controversy regarding interpretation of equity, a term that public discourse has used in vague and conflicting ways. Specifying social welfare functions that formally express different interpretations of equity will not eliminate disagreements, but it should clarify concepts and reduce the inconsistencies that afflict verbal communication.

econ.GN↗

arXiv2025

A Decision Theoretic Perspective on Artificial Superintelligence: Coping with Missing Data Problems in Prediction and Treatment Choice

Enormous attention and resources are being devoted to the quest for artificial general intelligence and, even more ambitiously, artificial superintelligence. We wonder about the implications for methodological research that aims to help decision makers cope with what econometricians call identification problems, inferential problems in empirical research that do not diminish as sample size grows. Of particular concern are missing data problems in prediction and treatment choice. Essentially all data collection intended to inform decision making is subject to missing data, which gives rise to identification problems. Thus far, we see no indication that the current dominant architecture of machine learning (ML)-based artificial intelligence (AI) systems will outperform humans in this context. In this paper, we explain why we have reached this conclusion and why we see the missing data problem as a cautionary case study in the quest for superintelligence more generally. We first discuss the concept of intelligence, focusing initially on some work by AI researchers, before presenting a decision-theoretic perspective that formalizes the connection between intelligence and identification problems. We next apply this perspective to two leading cases of missing data problems. Then we explain why we are skeptical that AI research is currently on a path toward machines doing better than humans at solving these identification problems.

econ.EM↗

arXiv2025

Utilitarian or Quantile-Welfare Evaluation of Social Welfare? With Application to Health Cost-Effectiveness Analysis

This paper considers quantile-welfare evaluation of social welfare as an alternative to utilitarian evaluation. Manski (1988) originally proposed and studied maximization of quantile utility as a model of individual decision making under uncertainty, juxtaposing it with maximization of expected utility. That paper's primary motivation was to exploit the fact that maximization of quantile utility requires only an ordinal formalization of utility, not a cardinal one. This paper transfers these ideas from analysis of individual decision making to analysis of social planning. We begin by summarizing basic theoretical properties of quantile welfare in general terms. We then turn attention to health cost-effectiveness analysis and consider measurement and econometric issues arising in that context. We propose a procedure to nonparametrically bound the quantile welfare of health states using data from binary-choice time-tradeoff (TTO) experiments of the type regularly performed by health economists.

econ.EM↗

arXiv2025

Comprehensive OOS Evaluation of Predictive Algorithms with Statistical Decision Theory

We argue that comprehensive out-of-sample (OOS) evaluation using statistical decision theory (SDT) should replace the current practice of K-fold and Common Task Framework validation in machine learning (ML) research on prediction. SDT provides a formal frequentist framework for performing comprehensive OOS evaluation across all possible (1) training samples, (2) populations that may generate training data, and (3) populations of prediction interest. Regarding feature (3), we emphasize that SDT requires the practitioner to directly confront the possibility that the future may not look like the past and to account for a possible need to extrapolate from one population to another when building a predictive algorithm. For specificity, we consider treatment choice using conditional predictions with alternative restrictions on the state space of possible populations that may generate training data. We discuss application of SDT to the problem of predicting patient illness to inform clinical decision making. SDT is simple in abstraction, but it is often computationally demanding to implement. We call on ML researchers, econometricians, and statisticians to expand the domain within which implementation of SDT is tractable.

econ.EM↗

arXiv2024

Using Ordinal Voting to Compare the Utilitarian Welfare of a Status Quo and A Proposed Policy: A Simple Nonparametric Analysis

The relationship of policy choice by majority voting and by maximization of utilitarian welfare has long been discussed. I consider choice between a status quo and a proposed policy when persons have interpersonally comparable cardinal utilities taking values in a bounded interval, voting is compulsory, and each person votes for a policy that maximizes utility. I show that knowledge of the attained status quo welfare and the voting outcome yields an informative bound on welfare with the proposed policy. The bound contains the value of status quo welfare, so the better utilitarian policy is not known. The minimax-regret decision and certain Bayes decisions choose the proposed policy if its vote share exceeds the known value of status quo welfare. This procedure differs from majority rule, which chooses the proposed policy if its vote share exceeds 1/2.

econ.TH↗

arXiv2024

Using Total Margin of Error to Account for Non-Sampling Error in Election Polls: The Case of Nonresponse

The potential impact of non-sampling errors on election polls is well known, but measurement has focused on the margin of sampling error. Survey statisticians have long recommended measurement of total survey error by mean square error (MSE), which jointly measures sampling and non-sampling errors. We think it reasonable to use the square root of maximum MSE to measure the total margin of error (TME). Measurement of TME should encompass both sampling error and all forms of non-sampling error. We suggest that measurement of TME should be a standard feature in the reporting of polls. To provide a clear illustration, and because we believe the exceedingly low response rates commonly obtained by election polls to be a particularly worrisome source of potential error, we demonstrate how to measure the potential impact of nonresponse using the concept of TME. We first show how to measure TME when a pollster lacks any knowledge of the candidate preferences of nonrespondents. We then extend the analysis to settings where the pollster has partial knowledge that bounds the preferences of non-respondents. In each setting, we derive a simple poll estimate that approximately minimizes TME, a midpoint estimate, and compare it to a conventional poll estimate.

econ.EM↗

arXiv2024

The Subtlety of Optimal Paternalism in a Population with Bounded Rationality

We study the subtlety of optimal paternalism when a utilitarian planner has the power to design a discrete choice set for a heterogeneous population with bounded rationality. We first consider the planning problem in abstraction. We show that the policy that most effectively constrains or influences choices depends multiplicatively on the preferences of the population and the choice probabilities conditional on preferences that measure the suboptimality of behavior. We then study two settings in which the planner may mandate an action or decentralize decision making. One setting supposes that individuals measure utility with additive random error and maximize mismeasured rather than actual utility. Then optimal planning requires knowledge of the distribution of measurement errors. The other setting studies binary treatment choice when the planner can mandate a treatment conditional on publicly observed personal covariates or can enable individuals to choose their own treatments conditional on private information. Here we focus on situations where bounded rationality takes the form of deviations between subjective and objective probabilities of uncertain outcomes. To illustrate, we consider clinical decision making in medicine. In toto, our cautionary analysis shows that determination of optimal policy requires the planner to possess extensive knowledge that is rarely available. We warn that research in behavioral public economics should avoid overoptimistic claims regarding the nature of optimal paternalistic policies. We argue that credible study of utilitarian planning should consider not only the population but also the planner to be boundedly rational.

econ.EM↗

arXiv2024

Identification and Statistical Decision Theory

Econometricians have usefully separated study of estimation into identification and statistical components. Identification analysis, which assumes knowledge of the probability distribution generating observable data, places an upper bound on what may be learned about population parameters of interest with finite sample data. Yet Wald's statistical decision theory studies decision making with sample data without reference to identification, indeed without reference to estimation. This paper asks if identification analysis is useful to statistical decision theory. The answer is positive, as it can yield an informative and tractable upper bound on the achievable finite sample performance of decision criteria. The reasoning is simple when the decision relevant parameter is point identified. It is more delicate when the true state is partially identified and a decision must be made under ambiguity. Then the performance of some criteria, such as minimax regret, is enhanced by randomizing choice of an action. This may be accomplished by making choice a function of sample data. I find it useful to recast choice of a statistical decision function as selection of choice probabilities for the elements of the choice set. Using sample data to randomize choice conceptually differs from and is complementary to its traditional use to estimate population parameters.

econ.EM↗

arXiv2023

Statistical Decision Theory Respecting Stochastic Dominance

The statistical decision theory pioneered by Wald (1950) has used state-dependent mean loss (risk) to measure the performance of statistical decision functions across potential samples. We think it evident that evaluation of performance should respect stochastic dominance, but we do not see a compelling reason to focus exclusively on mean loss. We think it instructive to also measure performance by other functionals that respect stochastic dominance, such as quantiles of the distribution of loss. This paper develops general principles and illustrative applications for statistical decision theory respecting stochastic dominance. We modify the Wald definition of admissibility to an analogous concept of stochastic dominance (SD) admissibility, which uses stochastic dominance rather than mean sampling performance to compare alternative decision rules. We study SD admissibility in two relatively simple classes of decision problems that arise in treatment choice. We reevaluate the relationship between the MLE, James-Stein, and James-Stein positive part estimators from the perspective of SD admissibility. We consider alternative criteria for choice among SD-admissible rules. We juxtapose traditional criteria based on risk, regret, or Bayes risk with analogous ones based on quantiles of state-dependent sampling distributions or the Bayes distribution of loss.

econ.EM↗

arXiv2023

Using Limited Trial Evidence to Credibly Choose Treatment Dosage when Efficacy and Adverse Effects Weakly Increase with Dose

In medical treatment and elsewhere, it has become standard to base treatment intensity (dosage) on evidence in randomized trials. Yet it has been rare to study how outcomes vary with dosage. In trials to obtain drug approval, the norm has been to specify some dose of a new drug and compare it with an established therapy or placebo. Design-based trial analysis views each trial arm as qualitatively different, but it may be highly credible to assume that efficacy and adverse effects (AEs) weakly increase with dosage. Optimization of patient care requires joint attention to both, as well as to treatment cost. This paper develops methodology to credibly use limited trial evidence to choose dosage when efficacy and AEs weakly increase with dose. I suppose that dosage is an integer choice t in (0, 1, . . . , T), T being a specified maximum dose. I study dosage choice when trial evidence on outcomes is available for only K dose levels, where K < T + 1. Then the population distribution of dose response is partially rather than point identified. The identification region is a convex polygon determined by linear equalities and inequalities. I characterize clinical and public-health decision making using the minimax-regret criterion. A simple analytical solution exists when T = 2 and computation is tractable when T is larger.

econ.EM↗

arXiv2022

Partial Identification of Personalized Treatment Response with Trial-reported Analyses of Binary Subgroups

Medical journals have adhered to a reporting practice that seriously limits the usefulness of published trial findings. Medical decision makers commonly observe many patient covariates and seek to use this information to personalize treatment choices. Yet standard summaries of trial findings only partition subjects into broad subgroups, typically into binary categories. Given this reporting practice, we study the problem of inference on long mean treatment outcomes E[y(t)|x], where t is a treatment, y(t) is a treatment outcome, and the covariate vector x has length K, each component being a binary variable. The available data are estimates of {E[y(t)|xk = 0], E[y(t)|xk = 1], P(xk)}, k = 1, . . . , K reported in journal articles. We show that reported trial findings partially identify {E[y(t)|x], P(x)}. Illustrative computations demonstrate that the summaries of trial findings in journal articles may imply only wide bounds on long mean outcomes. One can realistically tighten inferences if one can combine reported trial findings with credible assumptions having identifying power, such as bounded-variation assumptions.

econ.EM↗

arXiv2022

Inference with Imputed Data: The Allure of Making Stuff Up

Incomplete observability of data generates an identification problem. There is no panacea for missing data. What one can learn about a population parameter depends on the assumptions one finds credible to maintain. The credibility of assumptions varies with the empirical setting. No specific assumptions can provide a realistic general solution to the problem of inference with missing data. Yet Rubin has promoted random multiple imputation (RMI) as a general way to deal with missing values in public-use data. This recommendation has been influential to empirical researchers who seek a simple fix to the nuisance of missing data. This paper adds to my earlier critiques of imputation. It provides a transparent assessment of the mix of Bayesian and frequentist thinking used by Rubin to argue for RMI. It evaluates random imputation to replace missing outcome or covariate data when the objective is to learn a conditional expectation. It considers steps that might help combat the allure of making stuff up.

econ.EM↗

arXiv2022

Patient-Centered Appraisal of Race-Free Clinical Risk Assessment

Until recently, there has been a consensus that clinicians should condition patient risk assessments on all observed patient covariates with predictive power. The broad idea is that knowing more about patients enables more accurate predictions of their health risks and, hence, better clinical decisions. This consensus has recently unraveled with respect to a specific covariate, namely race. There have been increasing calls for race-free risk assessment, arguing that using race to predict patient outcomes contributes to racial disparities and inequities in health care. Writers calling for race-free risk assessment have not studied how it would affect the quality of clinical decisions. Considering the matter from the patient-centered perspective of medical economics yields a disturbing conclusion: Race-free risk assessment would harm patients of all races.

econ.EM↗