Fernandez de Arroyabe Arranz, Carlos (2022) The application of Machine Learning methods to the analysis of Business: A study of the implementation of the Circular Economy. PhD thesis, University of Essex.
Fernandez de Arroyabe Arranz, Carlos (2022) The application of Machine Learning methods to the analysis of Business: A study of the implementation of the Circular Economy. PhD thesis, University of Essex.
Fernandez de Arroyabe Arranz, Carlos (2022) The application of Machine Learning methods to the analysis of Business: A study of the implementation of the Circular Economy. PhD thesis, University of Essex.
Abstract
The thesis examines the application of different machine learning tools to the analysis of the implementation of circular economy in firms, to be able to better understand and solve the challenges these types of models pose for businesses, governments, and society. Particularly, this thesis studies how institutional pressures in different policy and business areas affect the development and promotion of circular economy models in firms, making special emphasis on the interaction of policies and the non-linearity and complementarity of the process. Hence, a combination of regression methods and machine learning (i.e., Artificial Neural Networks, K-means clusters, and Tree regression analysis) is used to analyse data from 870 companies in the European Union. The research is structured around three papers, which analyse three different key dimensions of the institutional environment of the company when developing a circular economy. That is, the effect of the typology of the institutional pressure, the economic actors (i.e. consumers and producers), and two economic activities (i.e. innovation and financial support). For this, the thesis brings together several perspectives of institutional theory (i.e., institutional pressures, institutional entrepreneurship, and institutional complexity) with stakeholder theory and dynamic capabilities theory. The combination of the three papers in the thesis shows that the application of machine learning tools has an important contribution in solving complex analytical questions involving multivariate non-linear relationships, complementarity, and interaction. Hence, an adequate combination of conventional regression analysis with machine learning can serve as an instrumental framework that helps increase the explanatory power of models suitable for the study of the circular economy. Moreover, the thesis contributes to the circular economy and institutional theory literature, particularly the extant literature on circular economy institutional pressures and policies, by better understanding and explaining their effect on circular economy models in firms, as well as providing interesting environmental policy and managerial implications.
Item Type: | Thesis (PhD) |
---|---|
Uncontrolled Keywords: | Machine Learning; Circular Economy; Business Analytics; Institutional Pressures |
Subjects: | H Social Sciences > H Social Sciences (General) |
Divisions: | Faculty of Social Sciences > Essex Business School |
Depositing User: | Carlos Fernandez De Arroyabe Arranz |
Date Deposited: | 09 Nov 2022 16:28 |
Last Modified: | 09 Nov 2022 16:28 |
URI: | http://repository.essex.ac.uk/id/eprint/33852 |
Available files
Filename: Thesis_Carlos F Arroyabe Arranz_1907719.pdf