The recent discovery of strong analogies between speculative markets and some well known physical phenomena and concepts, such as spin systems, universality, criticality and complexity has led to a growing interest of physicists in the dynamics of financial markets. Together with the development of these analogies between the statistical mechanics branch of physics and economics, the new field of study called econophysics is born. This field of research profits from the immense amount of data that nowadays are provided by different sources: from social networks to search engines. Following the procedure of recent studies, in this thesis we investigate the interplay between finance-related news and tweets and financial markets. In particular, we consider, in a period of 9 years, the Twitter-and-news volume of the 30 stock companies that form the Dow Jones Industrial Average (DJIA) index and, as a first attempt, we explore time-lagged cross-correlations and Granger-causality tests. However, the non-stationary and non-gaussian nature of financial data requires a different tool that can overcome the limits of linear statistics. We found this tool in information theory; allowing us to propose a novel approach based on a multivariate transfer entropy analysis.

The recent discovery of strong analogies between speculative markets and some well known physical phenomena and concepts, such as spin systems, universality, criticality and complexity has led to a growing interest of physicists in the dynamics of financial markets. Together with the development of these analogies between the statistical mechanics branch of physics and economics, the new field of study called econophysics is born. This field of research profits from the immense amount of data that nowadays are provided by different sources: from social networks to search engines. Following the procedure of recent studies, in this thesis we investigate the interplay between finance-related news and tweets and financial markets. In particular, we consider, in a period of 9 years, the Twitter-and-news volume of the 30 stock companies that form the Dow Jones Industrial Average (DJIA) index and, as a first attempt, we explore time-lagged cross-correlations and Granger-causality tests. However, the non-stationary and non-gaussian nature of financial data requires a different tool that can overcome the limits of linear statistics. We found this tool in information theory; allowing us to propose a novel approach based on a multivariate transfer entropy analysis.

MEASURING THE INFORMATION FLOW BETWEEN THE WEB AND STOCK MARKET VOLUMES: A MULTIVARIATE TRANSFER ENTROPY ANALYSIS

VICENTINI, GIULIO
2021/2022

Abstract

The recent discovery of strong analogies between speculative markets and some well known physical phenomena and concepts, such as spin systems, universality, criticality and complexity has led to a growing interest of physicists in the dynamics of financial markets. Together with the development of these analogies between the statistical mechanics branch of physics and economics, the new field of study called econophysics is born. This field of research profits from the immense amount of data that nowadays are provided by different sources: from social networks to search engines. Following the procedure of recent studies, in this thesis we investigate the interplay between finance-related news and tweets and financial markets. In particular, we consider, in a period of 9 years, the Twitter-and-news volume of the 30 stock companies that form the Dow Jones Industrial Average (DJIA) index and, as a first attempt, we explore time-lagged cross-correlations and Granger-causality tests. However, the non-stationary and non-gaussian nature of financial data requires a different tool that can overcome the limits of linear statistics. We found this tool in information theory; allowing us to propose a novel approach based on a multivariate transfer entropy analysis.
2021
MEASURING THE INFORMATION FLOW BETWEEN THE WEB AND STOCK MARKET VOLUMES: A MULTIVARIATE TRANSFER ENTROPY ANALYSIS
The recent discovery of strong analogies between speculative markets and some well known physical phenomena and concepts, such as spin systems, universality, criticality and complexity has led to a growing interest of physicists in the dynamics of financial markets. Together with the development of these analogies between the statistical mechanics branch of physics and economics, the new field of study called econophysics is born. This field of research profits from the immense amount of data that nowadays are provided by different sources: from social networks to search engines. Following the procedure of recent studies, in this thesis we investigate the interplay between finance-related news and tweets and financial markets. In particular, we consider, in a period of 9 years, the Twitter-and-news volume of the 30 stock companies that form the Dow Jones Industrial Average (DJIA) index and, as a first attempt, we explore time-lagged cross-correlations and Granger-causality tests. However, the non-stationary and non-gaussian nature of financial data requires a different tool that can overcome the limits of linear statistics. We found this tool in information theory; allowing us to propose a novel approach based on a multivariate transfer entropy analysis.
Complex Systems
Collective Dynamics
Information Theory
Social Network
Econphysics
File in questo prodotto:
File Dimensione Formato  
Vicentini_Giulio.pdf

accesso aperto

Dimensione 1.92 MB
Formato Adobe PDF
1.92 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/40223