Optimizing distancebased methods for large data s

Authors: Tobias Scholl Thomas Brenner

Publish Date: 2015/10/10

Volume: 17, Issue: 4, Pages: 333-351

PDF Link

Abstract

Distancebased methods for measuring spatial concentration of industries have received an increasing popularity in the spatial econometrics community However a limiting factor for using these methods is their computational complexity since both their memory requirements and running times are in mathcal On2 In this paper we present an algorithm with constant memory requirements and shorter running time enabling distancebased methods to deal with large data sets We discuss three recent distancebased methods in spatial econometrics the DOIndex by Duranton and Overman Rev Econ Stud 7241077–1106 2005 the Mfunction by Marcon and Puech J Econ Geogr 105745–762 2010 and the ClusterIndex by Scholl and Brenner Reg Stud aheadofprint1–15 2014 Finally we present an alternative calculation for the latter index that allows the use of data sets with millions of firms

Keywords: