The Use of Exhaustive Micro-Data Firm Databases for Economic Geography: The Issues of Geocoding and Usability in the Case of the Amadeus Database

Abstract

Economic geography has begun to explore the options involved in micro-data. New databases have become available and new techniques and an increase in computer power allow their treatment. However, two major issues impede the use of these datasets: the lack of geocoded spatial location and lack of exhaustivity in coverage. In this article, I explore the possibilities of using large micro-scale firm databases for economic geography in Europe. I show that current evolution in European official spatial data dissemination alows for geocoding of such databases using means that are accessible for researchers with minimal programming knowledge. For the specific case of the Amadeus database of the Bureau Van Dijk, I show that its limitations in terms of coverage have to be taken into acount, but do not hinder its use for analysis. Resulting maps show how the data allows to go further than classic databases such as the Eurostat Structural Business Statistics.

Big enterprise registration data imputation: Supporting spatiotemporal analysis of industries in China

“…In these studies, enterprise or firm data is widely used, including aggregated data, and micro enterprise data. As distinct from aggregated data, micro enterprise data allows users to analyze information at varying spatial levels or partitions, and provides much more fine-grained individual information, offering the potential for theoretical innovation in economic geography and regional studies that are invisible in aggregated data sets (Domenech, Lazzeretti, Molina, & Ruiz, 2011;Lennert, 2011).…”

“…Different from numerical data imputation, text data imputation can harness NLP for semantic analysis. For text data imputation, to label a text with predefined categories, text classification is required; to extract unambiguous location information and even accurate coordinates from georeferenced text data, location estimation and geocoding are often needed (Chen, David, & Yang, 2013;Lennert, 2011). For example, there is a need to classify, estimate and geocode text location for social media data (Barapatre, Meena, & Ibrahim, 2016;Ghahremanlou, Sherchan, & Thom, 2015;Krumm & Horvitz, 2015).…”

An Integrated Duranton and Overman Index and Local Duranton and Overman Index Framework for Industrial Spatial Agglomeration Pattern Analysis

“…However, traditional indexes are unable to accurately reflect agglomeration degrees, such as locational entropy, the Thiel index, spatial Gini coefficient, Herfindahl index, and EG index [13,15,[27][28][29][30][31]. This is mainly because these indexes were primarily designed for a fixed spatial scale [17], which makes them inevitably influenced by the zoning scheme of the administrative unit, i.e., the existence of the Modifiable Areal Unit Problem (MAUP) [22,32,33].…”