Gravity's Rainbow: A dynamic latent space model for the world trade network

Michael D. Ward, John S. Ahlquist, Arturas Rozenas

    Research output: Contribution to journalArticle

    Abstract

    The gravity model, long the empirical workhorse for modeling international trade, ignores network dependencies in bilateral trade data, instead assuming that dyadic trade is independent, conditional on a hierarchy of covariates over country, time, and dyad. We argue that there are theoretical as well as empirical reasons to expect network dependencies in international trade. Consequently, standard gravity models are empirically inadequate. We combine a gravity model specification with latent space networks to develop a dynamic mixture model for real-valued directed graphs. The model simultaneously incorporates network dependencies in both trade incidence and trade volumes. We estimate this model using bilateral trade data from 1990 to 2008. The model substantially outperforms standard accounts in terms of both in- and out-of-sample predictive heuristics. We illustrate the model's usefulness by tracking trading propensities between the USA and China.

    Original languageEnglish (US)
    Pages (from-to)95-118
    Number of pages24
    JournalNetwork Science
    Volume1
    Issue number1
    DOIs
    StatePublished - Apr 1 2013

    Fingerprint

    Space Simulation
    Gravitation
    world trade
    International trade
    China
    Incidence
    Dependency (Psychology)
    Directed graphs
    dyad
    heuristics
    incidence
    Specifications

    Keywords

    • gravity model
    • latent space
    • networks
    • trade

    ASJC Scopus subject areas

    • Communication
    • Social Psychology
    • Sociology and Political Science

    Cite this

    Gravity's Rainbow : A dynamic latent space model for the world trade network. / Ward, Michael D.; Ahlquist, John S.; Rozenas, Arturas.

    In: Network Science, Vol. 1, No. 1, 01.04.2013, p. 95-118.

    Research output: Contribution to journalArticle

    Ward, Michael D. ; Ahlquist, John S. ; Rozenas, Arturas. / Gravity's Rainbow : A dynamic latent space model for the world trade network. In: Network Science. 2013 ; Vol. 1, No. 1. pp. 95-118.
    @article{e09563106aeb4dddb3b737b822766401,
    title = "Gravity's Rainbow: A dynamic latent space model for the world trade network",
    abstract = "The gravity model, long the empirical workhorse for modeling international trade, ignores network dependencies in bilateral trade data, instead assuming that dyadic trade is independent, conditional on a hierarchy of covariates over country, time, and dyad. We argue that there are theoretical as well as empirical reasons to expect network dependencies in international trade. Consequently, standard gravity models are empirically inadequate. We combine a gravity model specification with latent space networks to develop a dynamic mixture model for real-valued directed graphs. The model simultaneously incorporates network dependencies in both trade incidence and trade volumes. We estimate this model using bilateral trade data from 1990 to 2008. The model substantially outperforms standard accounts in terms of both in- and out-of-sample predictive heuristics. We illustrate the model's usefulness by tracking trading propensities between the USA and China.",
    keywords = "gravity model, latent space, networks, trade",
    author = "Ward, {Michael D.} and Ahlquist, {John S.} and Arturas Rozenas",
    year = "2013",
    month = "4",
    day = "1",
    doi = "10.1017/nws.2013.1",
    language = "English (US)",
    volume = "1",
    pages = "95--118",
    journal = "Network Science",
    issn = "2050-1242",
    publisher = "Cambridge University Press",
    number = "1",

    }

    TY - JOUR

    T1 - Gravity's Rainbow

    T2 - A dynamic latent space model for the world trade network

    AU - Ward, Michael D.

    AU - Ahlquist, John S.

    AU - Rozenas, Arturas

    PY - 2013/4/1

    Y1 - 2013/4/1

    N2 - The gravity model, long the empirical workhorse for modeling international trade, ignores network dependencies in bilateral trade data, instead assuming that dyadic trade is independent, conditional on a hierarchy of covariates over country, time, and dyad. We argue that there are theoretical as well as empirical reasons to expect network dependencies in international trade. Consequently, standard gravity models are empirically inadequate. We combine a gravity model specification with latent space networks to develop a dynamic mixture model for real-valued directed graphs. The model simultaneously incorporates network dependencies in both trade incidence and trade volumes. We estimate this model using bilateral trade data from 1990 to 2008. The model substantially outperforms standard accounts in terms of both in- and out-of-sample predictive heuristics. We illustrate the model's usefulness by tracking trading propensities between the USA and China.

    AB - The gravity model, long the empirical workhorse for modeling international trade, ignores network dependencies in bilateral trade data, instead assuming that dyadic trade is independent, conditional on a hierarchy of covariates over country, time, and dyad. We argue that there are theoretical as well as empirical reasons to expect network dependencies in international trade. Consequently, standard gravity models are empirically inadequate. We combine a gravity model specification with latent space networks to develop a dynamic mixture model for real-valued directed graphs. The model simultaneously incorporates network dependencies in both trade incidence and trade volumes. We estimate this model using bilateral trade data from 1990 to 2008. The model substantially outperforms standard accounts in terms of both in- and out-of-sample predictive heuristics. We illustrate the model's usefulness by tracking trading propensities between the USA and China.

    KW - gravity model

    KW - latent space

    KW - networks

    KW - trade

    UR - http://www.scopus.com/inward/record.url?scp=84888059040&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84888059040&partnerID=8YFLogxK

    U2 - 10.1017/nws.2013.1

    DO - 10.1017/nws.2013.1

    M3 - Article

    VL - 1

    SP - 95

    EP - 118

    JO - Network Science

    JF - Network Science

    SN - 2050-1242

    IS - 1

    ER -