Robustness of agglomeration measures

Using the measures gathered by Beaudry & Schiffauerova and publishing those additionally gathered by GPS 2015, we show how the measures of agglomeration deviate from their true values in an analysis on data with a given DGP. We do so in three sections: in the first, we test the measures by themselves on DGPs based on each of the different measures, with multiple random iterations; in the second, we vary the DGP, and regress the results on the parameters in a setting that resembles meta-analysis; and in the third, we use actual data from the Netherlands to compare the outcomes of all variables without knowing the true DGP.

Status Work in progress, with prof. Frank van Oort. The work is partly based on previous discussions with Henri de Groot, Frank Neffke and Martijn Burger.

Leave a comment

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.