Konferenzpaper

Erschienen in: Multikonferenz Wirtschaftsinformatik (MKWI) 2016, Band 3

Jahr der Veröffentlichung: 2016

Seiten: 1787-1798

ISBN: 978-3-86360-132-4

Open Access Status: Gold

URL: https://nbn-resolving.org/urn:nbn:de:gbv:ilm1-2016100035

Konferenz: Multikonferenz Wirtschaftsinformatik (MKWI)

Abstract: 

The instant dissemination of news in financial markets results in unprecedented amounts of unstructured data. Such unstructured data can reveal interesting insights into the information processing of investors. However, research on information processing of a large set of potential covariates is challenging since knowledge about covariates is frequently scarce. As a remedy, this paper advocates a framework for Big Data analytics that studies the influence of news sentiment on prices when covariates are unknown. First, we apply a LASSO regularization to identify relevant covariates. We then integrate these into a linear regression, incorporate a news sentiment variable and evaluate the news reception. This paper demonstrates our research framework by utilizing the natural gas market, finding a positive effect of news sentiment on the natural gas price at a statistically significant level.

Harvard-Zitierstil: Alfano, S., Rapp, M., Pröllochs, N., Feuerriegel, S. and Neumann, D. (2016) Driven by News Tone? Understanding Information Processing when Covariates are Unknown: The Case of Natural Gas Price Movements, in Multikonferenz Wirtschaftsinformatik (MKWI) 2016, Band 3. Ilmenau: Universitätsverlag. pp. 1787-1798. https://nbn-resolving.org/urn:nbn:de:gbv:ilm1-2016100035

APA-Zitierstil: Alfano, S., Rapp, M., Pröllochs, N., Feuerriegel, S., & Neumann, D. (2016). Driven by News Tone? Understanding Information Processing when Covariates are Unknown: The Case of Natural Gas Price Movements. In Multikonferenz Wirtschaftsinformatik (MKWI) 2016, Band 3. (pp. 1787-1798). Universitätsverlag. https://nbn-resolving.org/urn:nbn:de:gbv:ilm1-2016100035