A Data-Driven Examination of Hotelling's Linear City Model

In his seminal work stability in competition, Hotelling developed a model for identifying the spatial equilibrium for two competing firms such that they maximize their market-share. He considered a linear area of fixed length and he showed that in this setting the two competing firms should be located side-by-side in the middle of the line. Hotelling's study has been then adopted and used to analyze and explain other phenomena in a variety of fields. However, the linear city model is purely theoretical, without any empirical validation. The goal of this study is to explore Hotelling's Law in its original space - i.e., that of firm competition - and identify possible adjustments needed to describe its application/validity in a non-linear city. In particular, we collect data from location-based social networks that include information for the number of customers in a venue and we compare them with the expectations from Hotelling's original law. Overall, we identify that at a large geographic scale there is correlation between the market-share and the inter-venue distance, which is consistent with the Hotelling's Law. However, as we zoom into smaller scales there are deviations from the expectations from Hotelling's law, possibly due to higher sensitivity to the necessary assumptions. Our findings enhance the literature on optimal location placement for a venue and can provide additional insights for owners in regards to the linear city model.

[1]  Alexander Kubis,et al.  Analysis of Location of Large-area Shopping Centres. A Probabilistic Gravity Model for the Halle–Leipzig Area , 2007 .

[2]  Ephraim S. Leibtag,et al.  Consumer shopping behavior: how much do consumers save? , 2009 .

[3]  V. Latora,et al.  Street Centrality and Densities of Retail and Services in Bologna, Italy , 2009 .

[4]  Oded Berman,et al.  Competitive facility location model with concave demand , 2007, Eur. J. Oper. Res..

[5]  Kerry D. Vandell,et al.  Retail Store Location and Market Analysis: A Review of the Research , 1992 .

[6]  Howard W. Smith,et al.  Streets, Malls, and Supermarkets , 2005 .

[7]  H. Hotelling.  Stability in Competition , 1929 .

[8]  Cecilia Mascolo,et al.  Geo-spotting: mining online location-based services for optimal retail store placement , 2013, KDD.

[9]  Pablo Jensen.  Network-based predictions of retail store commercial categories and optimal locations. , 2006, Physical review. E, Statistical, nonlinear, and soft matter physics.

[10]  Oded Berman,et al.  Efficient solution approaches for a discrete multi-facility competitive interaction model , 2009, Ann. Oper. Res..

[11]  V. Latora,et al.  Street Centrality and the Location of Economic Activities in Barcelona , 2012 .

[12]  M. Sheret.  Note on Methodology: The Coefficient of Variation , 1984, Comparative Education Review.

[13]  Ricardo J. G. B. Campello,et al.  Density-Based Clustering Based on Hierarchical Density Estimates , 2013, PAKDD.

[14]  Oded Berman,et al.  The generalized maximal covering location problem , 2002, Comput. Oper. Res..

[15]  Oded Berman,et al.  Competitive facility location and design problem , 2007, Eur. J. Oper. Res..

[16]  Pablo Jensen,et al.  Analyzing the Localization of Retail Stores with Complex Systems Tools , 2009, IDA.

[17]  Selim Balcisoy,et al.  A GIS-Based Optimization Framework for Competitive Multi-Facility Location-Routing Problem , 2010 .

[18]  S. Rosenthal,et al.  Evidence on the nature and sources of agglomeration economies , 2004 .

[19]  Mark S. Daskin,et al.  What you should know about location modeling , 2008 .

[20]  Rajiv D. Banker,et al.  Quality and Competition , 1998 .