Neural networks for regional employment forecasts: are the parameters relevant?

Abstract

In this paper, we present a review of various computational experiments concerning neural network (NN) models developed for regional employment forecasting. NNs are nowadays widely used in several fields because of their flexible specification structure. A series of NN experiments is presented in the paper, using two data sets on German NUTS-3 districts. Individual forecasts are computed by our models for each district in order to answer the following question: How relevant are NN parameters in comparison to NN structure? Comprehensive testing of these parameters is limited in the literature. Building on different specifications of NN models-in terms of explanatory variables and NN structures-we propose a systematic choice of NN learning parameters and internal functions by means of a sensitivity analysis. Our results show that different combinations of NN parameters provide significantly varying statistical performance and forecasting power. Finally, we note that the sets of parameters chosen for a given model specification cannot be light-heartedly applied to different or more complex models.

Mining urban land-use patterns from volunteered geographic information by means of genetic algorithms and artificial neural networks

“…The settings of the ANN and GA in this study are primarily chosen a priori. However, a detailed inspection and sensitivity analysis of the different settings bear potential for making the model more robust and yield better results (see Patuelli et al 2011). A fundamental problem of spatial data analysis on lattice data is the dependence of the results on the underlying scale and zoning (Openshaw 1984).…”

Systematization of labour market forecast models and lessons learned from forecasting

“…Minden modellt külön vizsgáltak Nyugat-és Kelet-Németországra. Az eredmények alapján azok a modellek működtek kisebb hibával, amelyek a shift-share analízist használták, vagyis azok, amelyek területi összehasonlításokat alkalmaztak (Patuelli et al [2011]). …”

“…Az egyik az, hogy nehezen formalizálható, azaz a modell elméleti alapokon történő specifikációja nem lehetséges. Sok esetben új modell leírására van szükség, amely az adott vizsgálandó jelenség függvénye, így nehéz elméleti hátteret kiépíteni a neurális hálók működésére (Patuelli et al [2011]). …”

Spatial Commuting Patterns of German Regional Labour Markets: A Sustainability Perspective

“…Nevertheless, there is a need for predicting future spatial and socio-economic patterns in a complex spatial system. This goal has led to more sophisticated modelling techniques which are also able to handle large-scale data (see, e.g., Gutiérrez-i-Puigarnau and Van Ommeren 2010; Nijkamp et al 2007;Patuelli et al 2011;Russo et al 2010;Travisi et al 2010). As well as new developments in spatial statistics and econometrics, scenario analysis has also become very popular as a means to understand complex systems (see Nijkamp et al , 2001.…”