IDEAS home Printed from https://ideas.repec.org/a/bpj/causin/v11y2023i1p53n1.html
   My bibliography  Save this article

Randomized graph cluster randomization

Author

Listed:
  • Ugander Johan

    (Management Science and Engineering, Stanford University, Stanford, CA 94305, California, United States)

  • Yin Hao

    (Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA 94305, California, United States)

Abstract

The global average treatment effect (GATE) is a primary quantity of interest in the study of causal inference under network interference. With a correctly specified exposure model of the interference, the Horvitz–Thompson (HT) and Hájek estimators of the GATE are unbiased and consistent, respectively, yet known to exhibit extreme variance under many designs and in many settings of interest. With a fixed clustering of the interference graph, graph cluster randomization (GCR) designs have been shown to greatly reduce variance compared to node-level random assignment, but even so the variance is still often prohibitively large. In this work, we propose a randomized version of the GCR design, descriptively named randomized graph cluster randomization (RGCR), which uses a random clustering rather than a single fixed clustering. By considering an ensemble of many different clustering assignments, this design avoids a key problem with GCR where the network exposure probability of a given node can be exponentially small in a single clustering. We propose two inherently randomized graph decomposition algorithms for use with RGCR designs, randomized 3-net and 1-hop-max, adapted from the prior work on multiway graph cut problems and the probabilistic approximation of (graph) metrics. We also propose weighted extensions of these two algorithms with slight additional advantages. All these algorithms result in network exposure probabilities that can be estimated efficiently. We derive structure-dependent upper bounds on the variance of the HT estimator of the GATE, depending on the metric structure of the graph driving the interference. Where the best-known such upper bound for the HT estimator under a GCR design is exponential in the parameters of the metric structure, we give a comparable upper bound under RGCR that is instead polynomial in the same parameters. We provide extensive simulations comparing RGCR and GCR designs, observing substantial improvements in GATE estimation in a variety of settings.

Suggested Citation

  • Ugander Johan & Yin Hao, 2023. "Randomized graph cluster randomization," Journal of Causal Inference, De Gruyter, vol. 11(1), pages 1-53, January.
  • Handle: RePEc:bpj:causin:v:11:y:2023:i:1:p:53:n:1
    DOI: 10.1515/jci-2022-0014
    as

    Download full text from publisher

    File URL: https://doi.org/10.1515/jci-2022-0014
    Download Restriction: no

    File URL: https://libkey.io/10.1515/jci-2022-0014?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    References listed on IDEAS

    as
    1. David Holtz & Sinan Aral, 2020. "Limiting Bias from Test-Control Interference in Online Marketplace Experiments," Papers 2004.12162, arXiv.org.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Nian Si, 2023. "Tackling Interference Induced by Data Training Loops in A/B Tests: A Weighted Training Approach," Papers 2310.17496, arXiv.org, revised Apr 2024.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Ozan Candogan & Chen Chen & Rad Niazadeh, 2024. "Correlated Cluster-Based Randomized Experiments: Robust Variance Minimization," Management Science, INFORMS, vol. 70(6), pages 4069-4086, June.
    2. Ali Goli & Anja Lambrecht & Hema Yoganarasimhan, 2024. "A Bias Correction Approach for Interference in Ranking Experiments," Marketing Science, INFORMS, vol. 43(3), pages 590-614, May.
    3. Hannah Li & Geng Zhao & Ramesh Johari & Gabriel Y. Weintraub, 2021. "Interference, Bias, and Variance in Two-Sided Marketplace Experimentation: Guidance for Platforms," Papers 2104.12222, arXiv.org.
    4. Ravi Bapna & Edward McFowland & Probal Mojumder & Jui Ramaprasad & Akhmed Umyarov, 2023. "So, Who Likes You? Evidence from a Randomized Field Experiment," Management Science, INFORMS, vol. 69(7), pages 3939-3957, July.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:bpj:causin:v:11:y:2023:i:1:p:53:n:1. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Peter Golla (email available below). General contact details of provider: https://www.degruyter.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.