Author

Jonathan Roth

28 papers · 1920–2026

1967peak 5 in 20212026

Frequent co-authors

Papers

showing 1–25 of 28 RSS

What's Trending in Difference-in-Differences? A Synthesis of the Recent Econometrics Literature

Jonathan Roth, Pedro H. C. Sant'Anna, Alyssa Bilinski +1 more

econ.EM · January 2022 · cited 1,695× · arXiv ↗ · more like this ·

This paper synthesizes recent advances in the econometrics of difference-in-differences (DiD) and provides concrete recommendations for practitioners. We begin by articulating a simple set of ``canonical'' assumptions under which the econometrics of DiD are well-understood. We then argue that recent advances in DiD methods can be broadly classified as relaxing some components of the canonical DiD setup, with a focus on $(i)$ multiple periods and variation in treatment timing, $(ii)$ potential violations of parallel trends, or $(iii)$ alternative frameworks for inference. Our discussion highlights the different ways that the DiD literature has advanced beyond the canonical model, and helps to clarify when each of the papers will be relevant for empirical work. We conclude by discussing some promising areas for future research.

A More Credible Approach to Parallel Trends

Ashesh Rambachan, Jonathan Roth

Statistics and Probability · The Review of Economic Studies · 15 Feb 2023 · cited 1,182× · DOI ↗ · more like this ·

Abstract This paper proposes tools for robust inference in difference-in-differences and event-study designs where the parallel trends assumption may be violated. Instead of requiring that parallel trends holds exactly, we impose restrictions on how different the post-treatment violations of parallel trends can be from the pre-treatment differences in trends (“pre-trends”). The causal parameter of interest is partially identified under these restrictions. We introduce two approaches that guarantee uniformly valid inference under the imposed restrictions, and we derive novel results showing that they have desirable power properties in our context. We illustrate how economic knowledge can inform the restrictions on the possible violations of parallel trends in two economic applications. We also highlight how our approach can be used to conduct sensitivity analyses showing what causal…

Pretest with Caution: Event-Study Estimates after Testing for Parallel Trends

Jonathan Roth

Statistics and Probability · American Economic Review Insights · 31 Aug 2022 · cited 698× · DOI ↗ · more like this ·

This paper discusses two important limitations of the common practice of testing for preexisting differences in trends (“ pre-trends”) when using difference-in-differences and related methods. First, conventional pre-trends tests may have low power. Second, conditioning the analysis on the result of a pretest can distort estimation and inference, potentially exacerbating the bias of point estimates and under-coverage of confidence intervals. I analyze these issues both in theory and in simulations calibrated to a survey of recent papers in leading economics journals, which suggest that these limitations are important in practice. I conclude with practical recommendations for mitigating these issues. (JEL A14, C23, C51)

Logs with zeros? Some problems and solutions

Jiafeng Chen, Jonathan Roth

econ.EM · December 2022 · cited 587× · arXiv ↗ · more like this ·

When studying an outcome $Y$ that is weakly-positive but can equal zero (e.g. earnings), researchers frequently estimate an average treatment effect (ATE) for a "log-like" transformation that behaves like $\log(Y)$ for large $Y$ but is defined at zero (e.g. $\log(1+Y)$, $\mathrm{arcsinh}(Y)$). We argue that ATEs for log-like transformations should not be interpreted as approximating percentage effects, since unlike a percentage, they depend on the units of the outcome. In fact, we show that if the treatment affects the extensive margin, one can obtain a treatment effect of any magnitude simply by re-scaling the units of $Y$ before taking the log-like transformation. This arbitrary unit-dependence arises because an individual-level percentage effect is not well-defined for individuals whose outcome changes from zero to non-zero when receiving treatment, and the units of the outcome…

When Is Parallel Trends Sensitive to Functional Form?

Jonathan Roth, Pedro H. C. Sant'Anna

econ.EM · October 2020 · cited 128× · arXiv ↗ · more like this ·

This paper assesses when the validity of difference-in-differences depends on functional form. We provide a novel characterization: the parallel trends assumption holds under all strictly monotonic transformations of the outcome if and only if a stronger ``parallel trends''-type condition holds for the cumulative distribution function of untreated potential outcomes. This condition for parallel trends to be insensitive to functional form is satisfied if and essentially only if the population can be partitioned into a subgroup for which treatment is effectively randomly assigned and a remaining subgroup for which the distribution of untreated potential outcomes is stable over time. These conditions have testable implications, and we introduce falsification tests for the null that parallel trends is insensitive to functional form.

Bias In, Bias Out? Evaluating the Folk Wisdom

Ashesh Rambachan, Jonathan Roth

cs.LG · September 2019 · cited 34× · arXiv ↗ · more like this ·

We evaluate the folk wisdom that algorithmic decision rules trained on data produced by biased human decision-makers necessarily reflect this bias. We consider a setting where training labels are only generated if a biased decision-maker takes a particular action, and so "biased" training data arise due to discriminatory selection into the training data. In our baseline model, the more biased the decision-maker is against a group, the more the algorithmic decision rule favors that group. We refer to this phenomenon as "bias reversal." We then clarify the conditions that give rise to bias reversal. Whether a prediction algorithm reverses or inherits bias depends critically on how the decision-maker affects the training data as well as the label used in training. We illustrate our main theoretical results in a simulation study applied to the New York City Stop, Question and Frisk dataset.

Interpreting Event-Studies from Recent Difference-in-Differences Methods

Jonathan Roth

econ.EM · January 2024 · cited 27× · arXiv ↗ · more like this ·

This note discusses the interpretation of event-study plots produced by recent difference-in-differences methods. I show that even when specialized to the case of non-staggered treatment timing, the default plots produced by software for several of the most popular recent methods do not match those of traditional two-way fixed effects (TWFE) event-studies. The plots produced by the new methods may show a kink or jump at the time of treatment even when the TWFE event-study shows a straight line. This difference stems from the fact that the new methods construct the pre-treatment coefficients asymmetrically from the post-treatment coefficients. As a result, visual heuristics for evaluating violations of parallel trends using TWFE event-study plots should not be immediately applied to those from these methods. I conclude with practical recommendations for constructing and interpreting…

War

Jonathan Roth

Anthropology · Cambridge University Press eBooks · 31 Dec 1920 · cited 15× · DOI ↗ · more like this ·

War was a major source of income for Hellenistic monarchs. A remarkable number of innovative military figures arose in Greece in the fourth century: Iphicrates and Chabrias of Athens, Epaminondas, Pelopidas and Pammenes of Thebes, Agesilaus of Sparta and Bolis of Crete, to name only a few. It was, however, Philip II of Macedon who drew many of the new elements of fourth-century warfare together, creating a formidable military machine. The Macedonian king fashioned a true combined arms force, streamlined the logistics of his army, and fashioned an up-to-date siege train. Perhaps most significantly, Philip wrought a new form of political organization, the Greek monarchy, able to direct this new military machine with a single will. Hellenistic strategic decisions were generally based on military circumstances and the desire to defeat the enemy and secure victory. Beginning with the…

Revised findings for "Procedural justice training reduces police use of force and complaints against officers"

George Wood, Tom R. Tyler, Andrew V. Papachristos +2 more

Health · RePEc: Research Papers in Economics · 21 Jan 2021 · cited 11× · DOI ↗ · more like this ·

Wood et al. (2020) studied the rollout of a procedural justice training program in the Chicago Police Department and found large and statistically significant impacts on complaints and sustained complaints against police officers and police use of force. This document describes a subtle statistical problem that led the magnitude of those estimates to be inflated. We then re-analyze the data using a methodology that corrects for this problem. The re-analysis provides less strong conclusions about the effectiveness of the training than the original study: although the point estimates for most outcomes and specifications are negative and of a meaningful magnitude, the confidence intervals typically include zero or very small effects. On the whole, we interpret the data as providing suggestive evidence that procedural justice training reduced the use of force, but no statistically…

Design-Based Uncertainty for Quasi-Experiments

Ashesh Rambachan, Jonathan Roth

econ.EM · August 2020 · cited 7× · arXiv ↗ · more like this ·

Design-based frameworks of uncertainty are frequently used in settings where the treatment is (conditionally) randomly assigned. This paper develops a design-based framework suitable for analyzing quasi-experimental settings in the social sciences, in which the treatment assignment can be viewed as the realization of some stochastic process but there is concern about unobserved selection into treatment. In our framework, treatments are stochastic, but units may differ in their probabilities of receiving treatment, thereby allowing for rich forms of selection. We provide conditions under which the estimands of popular quasi-experimental estimators correspond to interpretable finite-population causal parameters. We characterize the biases and distortions to inference that arise when these conditions are violated. These results can be used to conduct sensitivity analyses when there are…

Race and the Ivory Tower: An Antiracism Exercise for an Undergraduate Neuroscience Classroom.

Jonathan R. Roth, Cristin F. Gavin

Education · PubMed · 2021 · cited 4× · OpenAlex ↗ · more like this ·

This article details an antiracism exercise completed in an introductory undergraduate neuroscience class. Students completed an online pre-class multimedia module entitled "Race and the Ivory Tower" covering racism in science and medicine, the neuroscience behind bias, and the impact of race and racism on health outcomes. The module included two videos, one podcast, and a peer-reviewed journal article, alongside several optional additional resources written for both academic and lay audiences. After completing the module, students participated in an open-ended online discussion followed by an anonymous survey to elicit feedback on the exercise. As a continuation of the antiracism exercise, students researched and reported on the work of a Black or nonwhite Hispanic/Latino scientist for a final project later in the semester. Sixty-eight of 69 students participated in the discussion, and…

Efficient Estimation for Staggered Rollout Designs

Jonathan Roth, Pedro H. C. Sant'Anna

econ.EM · February 2021 · cited 3× · arXiv ↗ · more like this ·

We study estimation of causal effects in staggered rollout designs, i.e. settings where there is staggered treatment adoption and the timing of treatment is as-good-as randomly assigned. We derive the most efficient estimator in a class of estimators that nests several popular generalized difference-in-differences methods. A feasible plug-in version of the efficient estimator is asymptotically unbiased with efficiency (weakly) dominating that of existing approaches. We provide both $t$-based and permutation-test-based methods for inference. In an application to a training program for police officers, confidence intervals for the proposed estimator are as much as eight times shorter than for existing approaches.

Violence and the Roman Way of Warfare

Jonathan Roth

Anthropology · Cambridge University Press eBooks · 13 Mar 2020 · cited 3× · DOI ↗ · more like this ·

Scholars up until the middle of the twentieth century saw Roman warfare as restrained and disciplined. At that point the consensus changed to one that viewed it as fierce and bellicose. This view, in turn, has been challenged in the early twenty-first century, with the argument that Roman conflict was typical for ancient states. Rome’s rise from city-state to empire certainly involved considerable violence, but the available evidence cannot conclusive demonstrate either that it was particularly brutal and aggressive or that its military actions were ordinary for the period. Sources report that Roman battle was especially bloody, but this can be interpreted as a result of culture or of weaponry. We read of large numbers of civilians killed and enslaved, but such accounts need to be viewed critically and compared to the ancient norm. Additionally, the reality and nature of the imperial…

System theoretic approach to medical diagnosis

Igal Nevo, A. Guez, Fawad Ahmed +1 more

Artificial Intelligence · 10 Dec 2002 · cited 3× · DOI ↗ · more like this ·

A mathematical model for an adaptive expert system in anesthesia is presented. The concept of clusters that utilize clinical attributes in order to reduce the dimensionality of the patient's state-space is introduced. One goal of the model is to implement the existing categories and to identify and cluster categories as well as make the system adaptive to new and more optimal categories. Well-known techniques of pattern classification and cluster analysis are used on the measurable dataset to look for new categories or to readjust existing ones. Readjustment is required to optimize the existing categories to give the most efficient classification of the diseases.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>

Should We Adjust for the Test for Pre-trends in Difference-in-Difference Designs?

Jonathan Roth

econ.EM · April 2018 · cited 2× · arXiv ↗ · more like this ·

The common practice in difference-in-difference (DiD) designs is to check for parallel trends prior to treatment assignment, yet typical estimation and inference does not account for the fact that this test has occurred. I analyze the properties of the traditional DiD estimator conditional on having passed (i.e. not rejected) the test for parallel pre-trends. When the DiD design is valid and the test for pre-trends confirms it, the typical DiD estimator is unbiased, but traditional standard errors are overly conservative. Additionally, there exists an alternative unbiased estimator that is more efficient than the traditional DiD estimator under parallel trends. However, when in population there is a non-zero pre-trend but we fail to reject the hypothesis of parallel pre-trends, the DiD estimator is generally biased relative to the population DiD coefficient. Moreover, if the trend is…

(Empirical) Bayes Approaches to Parallel Trends

Soonwoo Kwon, Jonathan Roth

econ.EM · 18 Apr 2024 · cited 1× · arXiv ↗ · more like this ·

We consider Bayes and Empirical Bayes (EB) approaches for dealing with violations of parallel trends. In the Bayes approach, the researcher specifies a prior over both the pre-treatment violations of parallel trends $δ_{pre}$ and the post-treatment violations $δ_{post}$. The researcher then updates their posterior about the post-treatment bias $δ_{post}$ given an estimate of the pre-trends $δ_{pre}$. This allows them to form posterior means and credible sets for the treatment effect of interest, $τ_{post}$. In the EB approach, the prior on the violations of parallel trends is learned from the pre-treatment observations. We illustrate these approaches in two empirical applications.

Testing Mechanisms

Soonwoo Kwon, Jonathan Roth

econ.EM · April 2024 · cited 1× · arXiv ↗ · more like this ·

Economists are often interested in the mechanisms by which a treatment affects an outcome. We develop tests for the "sharp null of full mediation" that a treatment $D$ affects an outcome $Y$ only through a particular mechanism (or set of mechanisms) $M$. Our approach exploits connections between mediation analysis and the econometric literature on testing instrument validity. We also provide tools for quantifying the magnitude of alternative mechanisms when the sharp null is rejected: we derive sharp lower bounds on the fraction of individuals whose outcome is affected by the treatment despite having the same value of $M$ under both treatments (``always-takers''), as well as sharp bounds on the average effect of the treatment for such always-takers. An advantage of our approach relative to existing tools for mediation analysis is that it does not require stringent assumptions about how…

Simple Standard Errors for Complicated (Quasi-)Experimental Designs

Sylvia Klosin, Jonathan Roth

econ.EM · 8 Oct 2026 · cited 0× · arXiv ↗ · more like this ·

We develop standard errors for estimates of causal effects in settings where the distribution of treatment exposures is known but potentially exhibits complicated dependence patterns across units. Such dependence arises, for example, in settings with spillovers/interference across units or when the treatment assignment is spatially correlated. Our variance estimator generalizes several properties of the Neyman variance estimator for completely randomized experiments to general designs: it is unbiased for an upper bound on the variance from the design-based perspective; the conservativeness of this upper bound depends on the heterogeneity of treatment effects; and it is also valid under certain forms of super-population sampling of the potential outcomes. Our estimator reduces to standard approaches in simple designs, such as the Neyman variance estimator in completely randomized…

Development of small‐molecule tau‐SH3 interaction inhibitors that prevent amyloid‐β toxicity

Jonathan R. Roth, Travis Rush, Derian A. Pugh +9 more

Computational Theory and Mathematics · Alzheimer s & Dementia · December 2021 · cited 0× · DOI ↗ · more like this ·

Abstract Background Alzheimer’s disease (AD) is the leading cause of dementia and lacks disease‐modifying treatments. Tau‐based therapies are attractive and Tau reduction prevents amyloid‐induced dysfunction in preclinical AD models. Tau reduction also prevents amyloid‐independent dysfunction in diverse disease contexts, suggesting that strategies exploiting the mechanisms underlying Tau reduction may extend beyond AD. Tau’s interaction with SH3 domain containing proteins likely underlie mechanisms of amyloid‐induced dysfunction, so we targeted their interaction as a therapeutic strategy. Method We previously designed a peptide Tau‐SH3 interaction inhibitor based on the region of Tau that binds FynSH3, and determined that it prevented amyloid‐induced dysfunction in primary neurons. We next completed a high‐throughput screen to identify compounds that inhibit Tau‐Fyn interaction. Here…

An Outcome Test of Discrimination for Ranked Lists

Jonathan Roth, Guillaume Saint-Jacques, YinYin Yu

econ.EM · 15 Nov 2021 · cited 0× · arXiv ↗ · more like this ·

This paper extends Becker (1957)'s outcome test of discrimination to settings where a (human or algorithmic) decision-maker produces a ranked list of candidates. Ranked lists are particularly relevant in the context of online platforms that produce search results or feeds, and also arise when human decisionmakers express ordinal preferences over a list of candidates. We show that non-discrimination implies a system of moment inequalities, which intuitively impose that one cannot permute the position of a lower-ranked candidate from one group with a higher-ranked candidate from a second group and systematically improve the objective. Moreover, we show that that these moment inequalities are the only testable implications of non-discrimination when the auditor observes only outcomes and group membership by rank. We show how to statistically test the implied inequalities, and validate our…

Fred K. Drogula.Commanders and Command in the Roman Republic and Early Empire.Frederik J. Vervaet.The High Command in the Roman Republic: The Principle of thesummum imperium auspiciumquefrom 509 to 19 BCE.

Jonathan Roth

Anthropology · The American Historical Review · October 2016 · cited 0× · DOI ↗ · more like this ·

Despite the word “Command” being present in both titles, neither book under review addresses command culture, command and control, nor how the Roman high command dealt with logistics and intelligence (either tactical or strategic), nor how the Roman election of commanders affected warfare, nor the Senate’s role as a sort of combination General Staff and Ministry of War. There is little attempt to compare Rome’s command system to others, either ancient or modern. Neither book cites any specialist literature on command or military theory. In short, these are not military histories at all. Rather they are technical studies in Roman Republican constitutional law. Both Fred K. Drogula and Frederik J. Vervaet take as their starting point the work of the nineteenth-century German historian of Rome, Theodor Mommsen. In Mommsen’s model of Rome’s constitutional law (its Staatsrecht), each of the…

Multithreading and interprocessor communication in a dual-issue pipelined processor

Naraig Manjikian, Jonathan Roth

Hardware and Architecture · June 2008 · cited 0× · DOI ↗ · more like this ·

This paper describes multithreading and interprocessor communication support in a dual-issue pipelined 32-bit processor for prototyping single-chip multiprocessors in programmable logic. Multithreading support includes multiple register contexts and instructions for thread management. Interprocessor communication support includes a ring network interface embedded in the pipelined datapath with instructions for sending and receiving data through the interface. Synthesis results are presented for a multiprocessor system in an Altera Stratix chip, demonstrating that hardware support for eight threads constitutes 18% of the logic in each processor and the ring interface constitutes less than 3% of the logic.

Performance enhancement and high-level specification of a pipelined processor in programmable logic

Naraig Manjikian, Jonathan Roth

Hardware and Architecture · August 2007 · cited 0× · DOI ↗ · more like this ·

This paper describes the performance enhancement of a custom pipelined processor in programmable logic, and the description of the enhanced processor with the higher- level specification syntax of a custom software tool for system implementation in hardware. Faster operation of the processor is enabled with a revised clocking scheme for the embedded memory blocks that are used for the register file and the caches. A higher-level implementation of the enhanced processor is specified with a custom software tool called QUI 2 VER (Queen's University Interface Integrator in VHDL for Education and Research) that generates VHDL output. Synthesis results obtained with Altera Quartus II for a Stratix chip characterize the modest overhead for the datapath modifications to support the revised…

Roth on Rosenstein, 'Rome at War: Farms, Families and Death in the Middle Republic'

Jonathan Roth

Sociology and Political Science · San José State University ScholarWorks (San Jose State University) · 2005 · cited 0× · OpenAlex ↗ · more like this ·

New Perspectives on Rome's Farmer-SoldiersRome at War is a traditional monograph, which is very much a good thing.In this age of obfuscating postmodernizing, it is refreshing to see a scholar challenge a generally accepted thesis, systematically and creatively question the evidence and reason supporting it, and then propose a more viable model.Rosenstein's work is characterized by clear writing and an insistence on engaging all the evidence, even (in fact, especially) that which might undermine his own thesis.More historians need to get back to this tried and true method.However, that is not to say Rome at War is oldfashioned.The core of the argument revolves around a sophisticated use of recent advances in demographics, and although a background in statistics is useful, Rosenstein does an excellent job of making the discussion clear.

Evaluating Counterfactual Policies Using Instruments

Michal Kolesár, José Luis Montiel Olea, Jonathan Roth

econ.EM · December 2025 · arXiv ↗ · more like this ·

We study settings in which a researcher has an instrumental variable (IV) and seeks to evaluate the effects of a counterfactual policy that alters treatment assignment, such as a directive encouraging randomly assigned judges to release more defendants. We develop a general and computationally tractable framework for computing sharp bounds on the effects of such policies. Our approach does not require the often tenuous IV monotonicity assumption. Moreover, for an important class of policy exercises, we show that IV monotonicity -- while crucial for a causal interpretation of two-stage least squares -- does not tighten the bounds on the counterfactual policy impact. We analyze the identifying power of alternative restrictions, including the policy invariance assumption used in the marginal treatment effect literature, and develop a relaxation of this assumption. We illustrate our…