Auteur / Autrice : | Philipp Koch |
Direction : | Cesar Hidalgo |
Type : | Projet de thèse |
Discipline(s) : | Mathématiques et Applications |
Date : | Inscription en doctorat le 01/09/2021 |
Etablissement(s) : | Toulouse 1 |
Ecole(s) doctorale(s) : | École doctorale Mathématiques, informatique et télécommunications |
Partenaire(s) de recherche : | Laboratoire : TSE-R - Toulouse School of Economics - Recherche |
Mots clés
Résumé
The concept of economic complexity utilizes methods stemming from machine learning, in particular dimensionality reduction techniques, to analyze data describing the geography of economic activities. It allows to latently measure the amount of productive knowledge a country or region holds. Since the introduction of economic complexity in 2009, the idea has been applied to a variety of different data sources and several studies showed that economic complexity is a valid predictor of economic growth, income inequality, and greenhouse gas emissions. This thesis aims to, on the one hand, identify and fill existing gaps in the literature of the still comparatively young field of economic complexity. On the other hand, the idea of economic complexity has shown the power that lies in the intersection between methods originating from machine learning or artificial intelligence and economic data. Hence, another goal of this thesis is to create novel ideas in this interdisciplinary field.