Francesco is the co-founder and CEO of Brain, a company focused on the development of algorithms for trading strategies and investment decisions. He obtained his Ph.D. in Computational Methods applied to Quantum Physics from Uppsala University in 2010. He focused his career in solving complex computational problems in different sectors using a wide range of techniques, from density functional theory in solid state physics to the application of machine learning in different industrial sectors.
Machine Learning methods applied to Finance: Brain’s proprietary approaches for stock selection, clustering of funds and market scenarios.
We present Brain proprietary solutions to some common financial problems using machine learning techniques.
♦ The BSR signal is a daily stock ranking based on a supervised machine learning model that uses an ensemble of features related to market regimes, stock fundamentals, prices and volumes, calendar anomalies. The model can be customized with the specific investable universe, the rebalancing frequency and the investment style.
♦ Clustering of Market Scenarios: we use unsupervised machine learning to identify which days in the past are similar to current day, according to variables corresponding to a certain topic, e.g. financial stress.
♦ Clustering of Funds: Brain developed a platform for monitoring mutual funds. The platform uses unsupervised machine learning techniques to aggregate funds that show a similar behaviour according to a combination of various metrics. Among other things this approach enable the user to detect if a fund behaves too differently from other funds that are supposed to display similar characteristics.