Principal Component Analysis on ESG data

Abstract

The objective of the dissertation is to perform a principal component analysis on ESG data. The ESG database is provided by the ESG-Credit.eu project and contains information of 11.104 firms of all Europe. The research focuses on companies residing in French, Italy and Germany. The data present in the database are extracted from three different sources of information: Bloomberg, Thomson Reuters Eikon and CDP. The number of overall measures collected is 609 distributed among the three main pillars of the ESG score: Environment, Social and Governance. In the 20th century, ESG has played a crucial role in the financial environment, especially when it comes to the selection of investments and analysis of financial performance. It derives that the main purpose of the analysis in this research is to determine how clear and transparent is the information provided by the ESG data and which sub-components have the highest impact on each pillar (E, S, G) and on the overall score. By analysing the Principal Components, it would also be possible to determine whether the information contributed by each pillar is redundant or not. Finally, a principal component analysis of the data prior and post the “Paris Agreement on Climate Change” signed in 2016 is performed in order to evaluate the impact of the agreement on ESG data and a possible change of influence of the sub-components on the overall score.