Titolo della tesi: Emotions affect investors’ willingness to take risks during trading sessions. A machine learning & facial recognition experimental research
The study aims to delve into the analysis of human emotions in the trading decision-making process, investigating whether emotions (anger, disgust, fear, sadness, happiness, surprise) can contribute to changes in investors' risk aversion. The research is grounded in an "in-field" randomized control trial. The experiment is conducted on young Italian students, where facial expressions are recorded while they trade live on the American financial market. The study assesses whether an individual's emotional response correlates with subsequent risk-related decisions, measured by the number of transactions. The trial divides investors into two groups, inducing the treated investors to feel a higher level of fear by watching a fear-based video. Emotions are detected using facial recognition software capable of transforming micro-facial expressions into emotional states. As the main result of utilizing artificial intelligence, the research confirms that the treatment video used in the literature effectively elicits the emotion of fear. The experimental outcomes exhibit that when levels of fear and surprise increase, investors execute fewer transactions, indicating a heightened level of risk aversion.