Heterogeneity and Spatial Dependence of Regional Growth in the EU: A Recursive Partitioning Approach

Abstract

We use model‐based recursive partitioning to assess heterogeneity of growth and convergence processes based on economic growth regressions for 255 European Union NUTS2 regions from 1995 to 2005. Spatial dependencies are taken into account by augmenting the model‐based regression tree with a spatial lag. The starting point of the analysis is a human‐capital‐augmented Solow‐type growth equation similar in spirit to Mankiw et al. (1992, The Quarterly Journal of Economics, 107, 407–437). Initial GDP and the share of highly educated in the working age population are found to be important for explaining economic growth, whereas the investment share in physical capital is only significant for coastal regions in the PIIGS countries. For all considered spatial weight matrices recursive partitioning leads to a regression tree with four terminal nodes with partitioning according to (i) capital regions, (ii) non‐capital regions in or outside the so‐called PIIGS countries and (iii) inside the respective PIIGS regions furthermore between coastal and non‐coastal regions. The choice of the spatial weight matrix clearly influences the spatial lag parameter while the estimated slope parameters are very robust to it. This indicates that accounting for heterogeneity is an important aspect of modeling regional economic growth and convergence.

A simultaneous spatial autoregressive model for compositional data

“…For example partial least squares procedures for the Spatial LAG model have been proposed in Wang et al (2019) but for a single dependent variable. Similarly and with the same restriction, Spatial regression trees are developed for the LAG model in Wagner and Zeileis (2019). In a different direction, the aggregation of political parties in three blocks could be reconsidered.…”

Distributed Traffic Control Based on Road Network Partitioning Using Normalization Algorithm

“…In the study of traffic control subnetworks, static classification is mainly based on the traffic organization form of the road network. With the increasing number of motor vehicles, scholars began to study how to incorporate the dynamics of the traffic road network into the subnetwork classification algorithm in order to meet the demand of dynamic changes in traffic [45][46][47][48].…”

“…Qin [49] used simulation data to measure the temporal-spatial similarity of road operations in terms of road section density throughout the day and proposed a local road network subnetwork partitioning algorithm based on a macro fundamental diagram and spectral clustering algorithm. Wagner et al [45] developed two heuristic partitioning algorithms to find traffic subnetworks with uniform traffic levels using a defined mixed integer linear optimization model. Dong et al [50] combined the traffic dynamic characteristics of the road network, proposed a traffic subnetwork partitioning model based on the MFD theory, and proposed a key control zone identification method.…”

A Review of Software for Spatial Econometrics in R

“…A first attempt to insert a spatial lag model into a regression tree has been presented by Wagner and Zeileis [127], and is implemented in [128].…”