Nowadays, investors making financial decisions have access to plentiful information. In this paper, we investigate whether it is possible to use the information retrievable from a large number of online press articles to construct variables that help to predict the (conditional) stock market volatility. In particular, we use the Brexit referendum as an experiment and compare the predictive performance of a traditional GARCH model for the conditional variance of the FTSE 100 Index with a number of alternative specifications where measures of the tone and the proliferation of news related Brexit is used to augment the standard GARCH model. We find that most of the proposed news-augmented models outperform a traditional GARCH model both in-sample and out of sample.
Bellini, V., Guidolin, M., Pedio, M. (2020). Can Big Data Help to Predict Conditional Stock Market Volatility? An Application to Brexit. In Machine Learning, Optimization, and Data Science 6th International Conference, LOD 2020, Siena, Italy, July 19–23, 2020, Revised Selected Papers, Part I (pp.398-409). Springer Science and Business Media Deutschland GmbH [10.1007/978-3-030-64583-0_36].
Can Big Data Help to Predict Conditional Stock Market Volatility? An Application to Brexit
Pedio M.
2020
Abstract
Nowadays, investors making financial decisions have access to plentiful information. In this paper, we investigate whether it is possible to use the information retrievable from a large number of online press articles to construct variables that help to predict the (conditional) stock market volatility. In particular, we use the Brexit referendum as an experiment and compare the predictive performance of a traditional GARCH model for the conditional variance of the FTSE 100 Index with a number of alternative specifications where measures of the tone and the proliferation of news related Brexit is used to augment the standard GARCH model. We find that most of the proposed news-augmented models outperform a traditional GARCH model both in-sample and out of sample.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
