Guy Aridor

About Me

In Fall 2022 I will start as an Assistant Professor of Marketing at Northwestern University, Kellogg School of Management. I am mainly interested in understanding competition and regulatory issues in the digital economy as well as more broadly the societal impacts of digital technologies. My research uses tools from empirical industrial organization, experimental methodologies, and applied microeconomic theory in order to shed light on these questions.

Previously, I got my PhD in Economics at Columbia and a B.A. in computer science, pure/applied mathematics, and economics from Boston University. Before my PhD I spent a couple of years as a software engineer at HubSpot where I worked on a bunch of fun stuff, including product experimentation tools and scaling products.

Links: CV, Google Scholar, Github
Working Papers
  1. Drivers of Digital Attention: Evidence from a Social Media Experiment.
    • Short Abstract: I study demand for social media applications by conducting an experiment where I comprehensively monitor how participants spend their time and randomize access to applications. I characterize substitution patterns and estimate a demand model with inertia. I relate the findings to antitrust issues in these markets and argue that policies aimed at curbing digital addiction may be meaningful policy tools from an antitrust perspective.
    • Full Paper: PDF, Last Updated: December 2021
  2. The Effect of Privacy Regulation on the Data Industry: Empirical Evidence from GDPR. Joint work with Yeon-Koo Che, Tobias Salz.
    • Revision Requested, RAND Journal of Economics; Extended Abstract at EC'21
    • Short Abstract: We use novel data from the online travel industry to characterize the causal impact of GDPR on the data that firms can collect as well as their advertising revenues and ability to predict consumer behavior.
    • Full Paper: PDF, NBER, Last Updated: June 2022
  3. Competing Bandits: The Perils of Exploration under Competition. Joint work with Yishay Mansour, Alex Slivkins, Steven Wu.
    • Extended Abstract at EC'19
    • Short Abstract: We study the tension between exploration and competition and ask whether competition incentivizes the adoption of better exploration algorithms.
    • Full Paper: PDF, arXiv, Last Updated: June 2022
  4. Adaptive Efficient Coding: A Variational Auto-Encoder Approach. Joint work with Francesco Grechi, Michael Woodford.
    • Short Abstract: We study a model of neural coding that is based on the structure of a variational auto-encoder. We use the model to characterize how perception adapts as the underlying environment changes.
    • Full paper: PDF, biorXiv, Last Updated: May 2020
  1. Deconstructing the Filter Bubble: User Decision-Making and Recommender Systems. Joint work with Duarte Goncalves, Shan Sikdar.
    • Proceedings of the 14th ACM Conference on Recommender Systems (RecSys '20)
    • Short Abstract: We study a model of user decision making in the context of recommender systems. We show that user beliefs and risk-aversion levels are important for rationalizing existing empirical evidence and emphasize the usefulness of belief data in recommender system design and evaluation.
    • Full Paper: Publisher Version
  2. Recommenders' Originals: The Welfare Effects of the Dual Role of Platforms as Producers and Recommender Systems. Joint work with Duarte Goncalves.
    • Forthcoming, International Journal of Industrial Organization
    • Short Abstract: We characterize the welfare effects of the increased tendency for online platforms to both produce their own content and utilize recommender systems.
    • Full Paper: PDF, SSRN, Publisher Version
Work in Progress
  1. The Value of Recommender Systems: Decomposing the Informational and Discovery Gains. Joint work with Duarte Goncalves, Ruoyan Kong, Daniel Kluver, Joseph Konstan.
    • AEA RCT Pre-Registration
    • Short Abstract: We conduct a longitudinal field experiment on the movie recommendation platform MovieLens where we randomize the set of recommended movies and elicit beliefs about unseen movies. We use the data to decompose the influence that recommender systems have on consumption choices in terms of their informational and product discovery value.
  2. Shopping Alone: The Impact of The Decline of the American Mall. Joint work with Louise Guillouet, Howard Zhang.
Recorded Presentations
Press, Blogs, and Other Mentions