Spatial microsimulation modelling for retail market analysis at the small-area level

Kazumasa Hanaoka, Graham P. Clarke

    Research output: Contribution to journalArticlepeer-review

    19 Citations (Scopus)


    The purpose of this study is to construct a spatial microsimulation model known as the spatial microsimulation approach for retail market analysis (SMARMA) in order to analyse the retail market at the small-area level in Kusatsu City, Shiga Prefecture, Japan. In this study, we focus on examining the following issues. First, we attempt to create synthetic household microdata from a consumer questionnaire survey using both reweighting and imputation approaches. Second, we present the manner in which the results of the spatial microsimulation model are used for market analysis with regard to grocery stores. Market shares, turnover ranking and detailed consumer characteristics for selected stores are examined. In particular, the spatial distributions of households and their shopping behaviour are discussed in order to identify variations in consumer characteristics. As a result, this study shows that a spatial microsimulation model can generate detailed and reliable synthetic microdata from a consumer questionnaire survey. Besides, it is confirmed that this model is a highly relevant approach for implementing market analysis at the small-area level.

    Original languageEnglish
    Pages (from-to)162-187
    Number of pages26
    JournalComputers, Environment and Urban Systems
    Issue number2
    Publication statusPublished - 2007 Mar


    • Iterative proportional fitting
    • Kusatsu City
    • Retail market analysis
    • Simulated annealing
    • Spatial microsimulation
    • Synthetic household microdata

    ASJC Scopus subject areas

    • Geography, Planning and Development
    • Ecological Modelling
    • Environmental Science(all)
    • Urban Studies


    Dive into the research topics of 'Spatial microsimulation modelling for retail market analysis at the small-area level'. Together they form a unique fingerprint.

    Cite this