Author(s): Alberto Costantiello, Angelo Leogrande
In this article, we estimate the variable Ease of Doing Business-EDB in the context of Environmental, Social and Governance-ESG model. We use data from ESG World Bank dataset. We have used data from 193 countries in the period 2011-2020. The level of EDB is positively associated, among others, to “Individuals Using the Internet”, “Government Effectiveness”, “Cooling Degree Days”, and negatively associated to “Agriculture, Forestry, and Fishing, Value Added”, “Forest Area” and “Strength of Legal Rights Index”. Furthermore, we have applied a cluster analysis with the application of the k-Means algorithm optimized with the Elbow Method and we have found the presence of four clusters. Finally, we have proposed a confrontation among eight different machine- learning algorithms to predict the future value of EDB. We have found that Linear Regression is the best algorithm and that the level of EDB is expected to improve of 1.66% for the analysed countries.