An Economic Approach to Regulating Algorithms

Rambachan, Ashesh; Kleinberg, Jon; Mullainathan, Sendhil; Ludwig, Jens

doi:10.3386/w27111

Cited by 27 publications

(24 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much like previous research in other contexts (e.g. Lee and Solon, 2011; Rambachan and Roth, 2019), we find that the estimates are sensitive to this assumption; standard two-way fixed models in some cases lead to different qualitative conclusions than models with country-specific trends. Since identification in the models without trends is based on the more restrictive and more implausible assumption that behavioural outcomes across the world would have trended identically if not for the vaccination rollout, we prefer the model with trends.…”

Section: Econometric Methodssupporting

confidence: 80%

The Great Covid-19 Vaccine Rollout: Behavioural and Policy Responses

Auld

Toxvaerd

2021

NIER

View full text Add to dashboard Cite

Using daily data on vaccinations, disease spread and measures of social interaction from Google Mobility reports aggregated at the country level for 112 countries, we present estimates of behavioural responses to the global rollout of COVID-19 vaccines. We first estimate correlates of the timing and intensity of the vaccination rollout, finding that countries which vaccinated more of their population earlier strongly tended to be richer, whereas measures of the state of pandemic or its death toll up to the time of the initial vaccine rollout had little predictive ability after controlling for income. Estimates of models of social distancing and disease spread suggest that countries which vaccinated more quickly also experienced decreases in some measures of social distancing, yet also lower incidence of disease, and in these countries, policy-makers relaxed social distancing measures relative to countries which rolled out vaccinations more slowly.

show abstract

Section: Econometric Methodssupporting

confidence: 80%

The Great Covid-19 Vaccine Rollout: Behavioural and Policy Responses

Auld

Toxvaerd

2021

NIER

View full text Add to dashboard Cite

show abstract

“…We emphasize that this assumption may not be credible in some settings (see, e.g., Bilinski and Hatfield 2018; Kahn-Lang and Lang 2019; Rambachan and Roth 2019). The goal of our analysis, however, is to shed new light on a popular justification of the estimator as the DiD estimator under the simplest setting 1 .…”

Section: The Difference-in-differences Designmentioning

confidence: 97%

On the Use of Two-Way Fixed Effects Regression Models for Causal Inference with Panel Data

Imai¹,

Kim²

2020

Polit. Anal.

341

181

View full text Add to dashboard Cite

The two-way linear fixed effects regression (2FE) has become a default method for estimating causal effects from panel data. Many applied researchers use the 2FE estimator to adjust for unobserved unit-specific and time-specific confounders at the same time. Unfortunately, we demonstrate that the ability of the 2FE model to simultaneously adjust for these two types of unobserved confounders critically relies upon the assumption of linear additive effects. Another common justification for the use of the 2FE estimator is based on its equivalence to the difference-in-differences estimator under the simplest setting with two groups and two time periods. We show that this equivalence does not hold under more general settings commonly encountered in applied research. Instead, we prove that the multi-period difference-in-differences estimator is equivalent to the weighted 2FE estimator with some observations having negative weights. These analytical results imply that in contrast to the popular belief, the 2FE estimator does not represent a design-based, nonparametric estimation strategy for causal inference. Instead, its validity fundamentally rests on the modeling assumptions.

show abstract

“…Spillovers pose a general risk associated with scorebased targeting strategies. Prospective evaluations should be designed to account for them (Angelucci & Di Maro, 2015), and their mitigation should be considered together with other issues arising with the use of algorithms in healthcare (Obermeyer, Powers, Vogeli, & Mullainathan, 2019;Rambachan, Kleinberg, Mullainathan, & Ludwig, 2020).…”

Section: Resultsmentioning

confidence: 99%

Can targeting high-risk patients reduce readmission rates? Evidence from Israel

Shadmi

Zeltzer

Shir

et al. 2020

Journal of Applied Economics

View full text Add to dashboard Cite

We study a large intervention intended to reduce hospital readmission rates in Israel. Since 2012, readmission risk was calculated for patients aged 65 and older, and high-risk patients were flagged to providers upon admission and after discharge. Analyzing 171,541 admissions during 2009-2016, we find that the intervention reduced 30-day readmission rates by 5.9% among patients aged 65-70 relative to patients aged 60-64, who were not targeted by the intervention and for whom no risk-scores were calculated. The largest reduction, 12.3%, was among high-risk patients, though some of it may reflect substitution of attention away from patients with unknown high-risk at the point of care. Post-discharge follow-up encounters were significantly expedited. Estimated effects declined after incentives to reduce readmission rates were discontinued. The evidence demonstrates that informing providers about patient risk in real-time coupled with incentives to reduce readmissions can improve care continuity and reduce hospital readmissions.

show abstract

An Economic Approach to Regulating Algorithms

Cited by 27 publications

References 6 publications

The Great Covid-19 Vaccine Rollout: Behavioural and Policy Responses

The Great Covid-19 Vaccine Rollout: Behavioural and Policy Responses

On the Use of Two-Way Fixed Effects Regression Models for Causal Inference with Panel Data

Can targeting high-risk patients reduce readmission rates? Evidence from Israel

Contact Info

Product

Resources

About