Essays on Multivariate Modelling of Financial Markets Using Copula and Sentiment Networks

Download or read book Essays on Multivariate Modelling of Financial Markets Using Copula and Sentiment Networks written by Anastasija Tetereva and published by . This book was released on 2018 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Multivariate dependence structures play an important role in finance. The modelling and accurate prediction of multivariate financial time series is an important component of asset pricing and portfolio management. This doctoral thesis comprises three essays that address the question of multivariate dependencies using high-frequency data and innovative sources of information such as news analytics. These essays make complementary contributions to the field of financial econometrics and can be read independently of each other. The first essay focuses on the improvement of Value at Risk prediction based on highfrequency data. The novel concept of the realized hierarchical Archimedean copula is introduced. It is proposed estimating the structure and the parameters of the hierarchical Archimedean copula using the realized correlation matrix only. This approach allows one to estimate the multivariate distribution of daily returns based on intraday information. Moreover, the proposed estimator does not suffer from the curse of dimensionality. In this essay, the realized hierarchical Archimedean copula is applied to manage the risk of high-dimensional portfolios. The evidence of the superior forecasting power of our approach, compared to a set of existing models, is provided. The second essay investigates the role of news sentiment data in improving forecasts in financial econometrics. The objective of this paper is to answer the question regarding whether the class of stock-price-relevant news is wider than firm-specific announcements. For this purpose, causal links between news sentiments and excess returns are studied by means of an adaptive lasso. It is concluded that unexpected returns in the whole economy can be explained by news originating from the financial and energy sectors. In other words, the news spillover effects are dominating the direct effects of sectoral news. Therefore, including exogenous financial or energy sentim.