1. Applying separative non-negative matrix factorization to extra-financial data
- Author
-
Fogel, P, Geissler, C, Cotte, P, Luta, G, Advestis, Georgetown University [Washington] (GU), and Morizet, Nicolas
- Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI] ,FOS: Computer and information sciences ,[QFIN.CP] Quantitative Finance [q-fin]/Computational Finance [q-fin.CP] ,Computational Finance (q-fin.CP) ,Machine Learning (stat.ML) ,[STAT.ML] Statistics [stat]/Machine Learning [stat.ML] ,Clustering ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,Machine Learning ,FOS: Economics and business ,Quantitative Finance - Computational Finance ,[STAT.ML]Statistics [stat]/Machine Learning [stat.ML] ,Statistics - Machine Learning ,Dimension reduction ,Interpretability ,ESG data ,Features Engineering - Abstract
International audience; We present here an original application of the non-negative matrix factorization (NMF) method, for the case of extra-financial data. These data are subject to high correlations between co-variables, as well as between observations. NMF provides a much more relevant clustering of co-variables and observations than a simple principal component analysis (PCA). In addition, we show that an initial data separation step before applying NMF further improves the quality of the clustering.
- Published
- 2022
- Full Text
- View/download PDF