DTA

Corso Ordinario Secondo Livello

GELSI, FEDERICO

etd-10272022-094720

A comparative multimodal stock market prediction using ECB press conferences

Cl. Sc. Sociali - Scienze Economiche

SCIENZE ECONOMICHE E MANAGERIALI - SCIENZE ECONOMICHE E MANAGERIALI

Tutor Prof.ssa NUTI, SABINA
Relatore Prof. RAGUSA, GIUSEPPE
Membro Prof. PICCALUGA, ANDREA MARIO CUORE
Tutor Prof. TENUCCI, ANDREA
Tutor Prof. BELLINI, NICOLA
Membro Dott.ssa CANTARELLI, PAOLA
Membro Prof. IRALDO, FABIO

• European Central Bank
• Forecasting
• Multimodality
• Neural Networks

02/12/2022;

parziale

Predicting stock market movements is a challenging task that has been extensively researched in the literature. In this study, we have two objectives. The first is to explore the possibility of successfully predicting market surprises after the European Central Bank (ECB) press conferences through a multimodal approach that exploits audio and transcripts. The second, functional to the former, is creating the first Multimodal Aligned database of ECB Press Conferences (ECB - MAPC) to lay the foundations for further multimodal research on central bank communication. For the predictions, we use a two-stage approach based on a first extraction of context-dependent speech features and a second stage, where the actual forecast takes place. We then compare the predictive performance of multiple machine and deep learning models, finding that good levels of accuracy can be achieved. The results also show that multimodality is crucial for decreasing prediction error and that not all market indices are equally predictable.

Nome file	Dimensione
Ci sono 1 file riservati su richiesta dell'autore. Contatta l'autore