Zak Elamrani headshot

Zakaria El Amrani

PhD candidate in Economics

I am a fourth year PhD candidate in Economics at Northwestern University. My research interests are in econometrics and statistics, with a focus on the identification and estimation of high-dimensional models, network theory, and inference using bootstrap methods.

Interests

  • Econometrics
  • Statistics
  • Industrial Organization

Education

  • PhD in Economics (expected 2027)
    Northwestern University
  • MS in Economics (2023)
    Northwestern University
  • BA in Mathematics, summa cum laude (2018)
    Brandeis University

Working Papers

Zakaria El Amrani 2025

Bootstrap Bias Correction for Improved Coverage Accuracy in Nonparametric Inference

Zakaria El Amrani (Joint with Eric Auerbach) 2026

Partial-Identification of Networks from Short Panels

Zakaria El Amrani 2025

Estimation of Stochastic Block Networks from Panel Data

Network spillovers shape many economic outcomes, yet the underlying interaction network is often unobserved in the administrative and survey panels most commonly used in empirical work. This paper develops an econometric framework to estimate a stochastic block model (SBM) representation of a latent interaction network using only panel data on outcomes and covariates. We embed an SBM restriction into a canonical linear social-interactions (spatial autoregressive) panel model and show that the community structure (block memberships), block-to-block link intensities, and spillover parameters are jointly identified from reduced-form panel variation under mild rank and stability conditions. Exploiting the implied low-dimensional structure, we propose a computationally tractable estimator that (i) recovers an empirical “influence” object from the panel via regularized moment-based methods and (ii) uses spectral clustering and blockwise refinement to estimate community assignments and block connection parameters, with data-driven selection of the number of blocks. We establish consistency and provide asymptotic characterizations for both spillover parameters and community recovery under large-N, short-T asymptotics. Monte Carlo evidence shows that imposing stochastic block structure can substantially improve the precision and stability of recovered interaction patterns and spillover estimates relative to unrestricted link-by-link recovery. An empirical illustration demonstrates that the approach delivers economically interpretable communities and a parsimonious map of interactions when network data are unavailable.

Zakaria El Amrani (Joint with Anastasiia Evdokimova) 2026

Forward Looking Adoption Under Policy Uncertainty: Evidence From Healthcare

Work in Progress

Zakaria El Amrani Ongoing

Optimal Penalty Choice in High Dimensional Models

Zakaria El Amrani Ongoing

Sparse Network Estimation in Nodal Pricing: Propagation Multipliers and the Incidence of Battery Storage

Zakaria El Amrani (Joint with Jacint Enrich, Natalia Fabra and Mar Reguant) Ongoing

Quantifying the Solar Rebound with Smart Meter Data: The Role of Co-Adoptions

Publications (other)

Zakaria El Amrani (Joint with Kaushik Ghosh et al.) 2025
Health Services Research

Integrated health systems and medical care quality during the COVID-19 pandemic

Objective: To examine differences between patients treated in integrated systems of care and patients treated outside of such systems during the COVID pandemic in the use of primary and preventive care, emergency services, inpatient services, and mortality.

Data sources and study setting: Data are used from all enrollees in traditional Medicare aged 66 and older.

Study design: Difference-in-differences estimates are calculated from the pre-COVID time period (January 2019-February 2020) to the initial COVID time period (March-May 2020) and the ongoing COVID time period (June 2020-December 2021) for patients treated by primary care physicians working in a health system versus not, and by the type of health system.

Data collection/extraction methods: Medicare claims data are used to measure monthly claims for office and telehealth visits, mammography, colon cancer screening, inpatient/emergency department visits, and death. Patients are assigned to primary care physicians using common algorithms. Physician membership in a health system is determined from a previously generated dataset.

Principal findings: Relative to the pre-COVID period, patients treated in health systems fared no better in maintaining primary care access than patients treated outside of such systems (DID estimate on receipt of office care or telehealth visit = −4%; p < 0.001). In the ongoing COVID time period, non-COVID mortality rose by less in health systems (DID estimate = −0.9%; p < 0.001) and health system patients experienced a greater decline in the use of the emergency department (DID estimate = −1.2%; p < 0.001) and emergency/urgent inpatient care for non-COVID conditions less (DID estimate = −0.7%; p < 0.001).

Conclusion: Health systems were associated with reduced occurrence of death and adverse medical events, although the effect magnitudes are modest. This reduction appears unrelated to the use of primary care and should be considered in the context of our evolving understanding of the advantages and disadvantages of health systems.

Keywords: COVID; Medicare; adverse medical events; death; health system and provider database; health systems.

Zakaria El Amrani (Joint with Nancy Beaulieu et al.) 2023
Journal of the American Medical Association

Organization and Performance of US Health Systems

Importance Health systems play a central role in the delivery of health care, but relatively little is known about these organizations and their performance.

Objective To (1) identify and describe health systems in the United States; (2) assess differences between physicians and hospitals in and outside of health systems; and (3) compare quality and cost of care delivered by physicians and hospitals in and outside of health systems.

Evidence Review Health systems were defined as groups of commonly owned or managed entities that included at least 1 general acute care hospital, 10 primary care physicians, and 50 total physicians located within a single hospital referral region. They were identified using Centers for Medicare & Medicaid Services administrative data, Internal Revenue Service filings, Medicare and commercial claims, and other data. Health systems were categorized as academic, public, large for-profit, large nonprofit, or other private systems. Quality of preventive care, chronic disease management, patient experience, low-value care, mortality, hospital readmissions, and spending were assessed for Medicare beneficiaries attributed to system and nonsystem physicians. Prices for physician and hospital services and total spending were assessed in 2018 commercial claims data. Outcomes were adjusted for patient characteristics and geographic area.

Findings A total of 580 health systems were identified and varied greatly in size. Systems accounted for 40% of physicians and 84% of general acute care hospital beds and delivered primary care to 41% of traditional Medicare beneficiaries. Academic and large nonprofit systems accounted for a majority of system physicians (80%) and system hospital beds (64%). System hospitals were larger than nonsystem hospitals (67% vs 23% with >100 beds), as were system physician practices (74% vs 12% with >100 physicians). Performance on measures of preventive care, clinical quality, and patient experience was modestly higher for health system physicians and hospitals than for nonsystem physicians and hospitals. Prices paid to health system physicians and hospitals were significantly higher than prices paid to nonsystem physicians and hospitals (12%–26% higher for physician services, 31% for hospital services). Adjusting for practice size attenuated health systems differences on quality measures, but price differences for small and medium practices remained large.

Conclusions and Relevance In 2018, health system physicians and hospitals delivered a large portion of medical services. Performance on clinical quality and patient experience measures was marginally better in systems but spending and prices were substantially higher. This was especially true for small practices. Small quality differentials combined with large price differentials suggests that health systems have not, on average, realized their potential for better care at equal or lower cost.

Contact

zakaria.elamrani@kellogg.northwestern.edu
224-296-9525
2211 Campus Dr, Evanston, IL 60208