Title: News and intraday jumps: A big data approach
Authors: Massimiliano Caporin - University of Padova (Italy) [presenting]
Francesco Poli - University of Padova (Italy)
Abstract: The aim is to study how news provokes intraday price jumps in the S\&P 100 constituents. We build high-frequency indicators that go beyond the mere release of news and investigate their association with jumps by applying penalised logistic regression and by dealing with the rare nature of jumps with appropriate techniques. Relevant causes of jumps are found to be EPS, rate decisions, bad macro-news and company-specific news with specific topics or negative sentiment. Market players sometimes act before the public release of information. Finally, we find that news influences the economic significance of jumps in terms of returns predictability and volatility persistence.