Malerba, Federico (2018) Machine learning for learning trader's behaviour. [Laurea triennale]
Per questo documento il full-text online non disponibile.
In this report we present and discuss an attempt to model a trader’s behaviour; the aimis not to necessarily make profit or maximize revenue, but rather to create a machine that understands how a trader’s decision-making process works. The ideal algorithm should determine when the trader would short or long on a financial market good; this should be achieved just by seeing past behaviour of the trader. To simplify the problem, we worked under the assumption that the trader makes decisions based solely on market-related information and not on any type of news coming from the outside world; the trader we’ve been in contact with - who was responsible for labelling the data - has told us that this is the way in which he operates.
Solo per lo Staff dell Archivio: Modifica questo record