Methodological reasoning for determining optimal economic size of regions: a Multi-Layer Perceptron approach

Omar Benida (), Khalil Allali (), Hassan Ramou (), Aziz Fadelaoui () and Fayssal Fadili ()
Additional contact information
Omar Benida: Agronomic and Veterinary Institute Hassan II
Khalil Allali: National School of Agriculture of Meknes
Hassan Ramou: University- Institute of African, Euro-Mediterranean and Iberoamerican Studies
Aziz Fadelaoui: Regional Center for Agronomic Research of Meknes
Fayssal Fadili: School of Information Sciences

SN Business & Economics, 2025, vol. 5, issue 6, 1-25

Abstract: Abstract This article is intended as a methodological contribution to reasoning about the optimal economic size of a region. Determining this size enables public authorities to act to reduce economic inequalities between regions. However, econometric methods based on panel regressions are largely unaware of recent rapid developments in machine learning methods. This article proposes a predictive model based on the Multi-Layer Perceptron—non-linary regression to determine the optimal economic size of a region. The eight out of twenty variables selected to determine the optimal economic size of a region were statistically analyzed using SPSS before being introduced into the model. The model revealed a very low loss of around 0.0303, and a val_loss of 0.0527. This confirmed the good performance of the model adopted. The data prediction was obtained through an unconstrained optimization where all regions converge towards the average Gross Domestic Product and a simulation based on the Morocco's new development model to be adopted in June 2021 guidelines stipulating an average growth of 6% by 2035. The originality of this approach lies in the combination of economic, demographic, and environmental dimensions to determine the relevant variables of economic development. It also relies on the use of predictive modeling powered by Artificial Intelligence, in particular machine learning. The direct implications the results of this empirical approach are likely to enable researchers and doctoral students working on this theme of regionalization and economic growth to master the prediction of other socio-economic and political/governance variables with good precision.

Keywords: Economic growth; Regional division; Deep learning; Model MLP; Optimal economic size; Prediction (search for similar items in EconPapers)
JEL-codes: R10 R11 R12 (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s43546-024-00781-9 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:snbeco:v:5:y:2025:i:6:d:10.1007_s43546-024-00781-9

Ordering information: This journal article can be ordered from
https://www.springer.com/journal/43546

DOI: 10.1007/s43546-024-00781-9

Access Statistics for this article

SN Business & Economics is currently edited by Gino D'Oca

More articles in SN Business & Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().