Available via license: CC BY
Content may be subject to copyright.
A preview of the PDF is not available
Macroeconomic News Sentiment: Enhanced Risk Assessment for Sovereign Bonds
Abstract and Figures
We enhance the modelling and risk assessment of sovereign bond spreads by taking into account quantitative information gained from macro-economic news sentiment. We investigate sovereign bonds spreads of five European countries and improve the prediction of spread changes by incorporating news sentiment from relevant entities and macro-economic topics. In particular, we create daily news sentiment series from sentiment scores as well as positive and negative news volume and investigate their effects on yield spreads and spread volatility. We conduct a correlation and rolling correlation analysis between sovereign bond spreads and accumulated sentiment series and analyse changing correlation patterns over time. Market regimes are detected through correlation series and the impact of news sentiment on sovereign bonds in different market circumstances is investigated. We find best-suited external variables for forecasts in an ARIMAX model set-up. Error measures for forecasts of spread changes and volatility proxies are improved when sentiment is considered. These findings are then utilised to monitor sovereign bonds from European countries and detect changing risks through time.
Figures - available via license: CC BY
Content may be subject to copyright.
... Within commodities, Chatrath et al. (2012) looked at the impact of macroeconomic news on the price of crude oil, while (Smales 2014) investigated news sentiment in the gold futures market. Erlwein-Sayer (2018) has reported use of macroeconomic news for improved prediction of sovereign bond spreads and Ho et al. (2017) explores the role of news in China's foreign exchange market. Furthermore, Yang et al. (2015) study the relationship between news sentiment and cumulative returns of the entire market by proposing two methods for measuring the abnormal levels of news sentiment. ...
In this work we predict changes in the Gross Domestic Product (GDP) of China using dynamic factor models. We report results of 3- and 6-months ahead forecasts, where we use 124\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$124$$\end{document} predictors from various sources and dates ranging from 2000 through 2017. Our analysis includes China specific macroeconomic time series data and a large number of predictor variables. We follow the latest state of the art, as outlined by, Stock and Watson (in: Handbook of macroeconomics vol 2, Elsevier, pp 415–525, 2016) who use principal component analysis (PCA) to reduce number of variables and apply dynamic factor model (DFM) to make predictions. The results suggest that including news sentiment significantly improves forecasts and this approach outperforms univariate autoregression. The contributions of this paper are two fold, namely, the use of news to improve forecasts and superior forecast of China’s GDP.
For three countries of similar economic characteristics, I ratify previous studies of the impact of fundamental macroeconomic and foreign exchange variables influencing country risk, as captured by the Emerging Market Bond Index (EMBI). I contribute to existing research, first by calculating a proxy of risk I call endogenous risk that analyzes the quarterly variability of economic activity, and second, by calculating a variable of sentiment from Twitter activity. I gauge the impact of both on the country risk metric in addition to variables in existing research about the determinants of country risk. Foreign exchange variables are the most significant determinants of risk for the countries of Colombia and Peru, which actively manage their currency, while Ecuador’s country risk is mostly affected by endogenous risk and macroeconomic fundamentals.
Monitoring the default risk of banks is highly relevant for companies when they enter into derivative contracts with banks to hedge risks. These contracts are subject to the counterparty risk of the bank defaulting. In credit risk management, market-based measures such as credit default swap (CDS) spreads are used to continuously assess the credit risk of counterparties. At the same time, CDS spreads are difficult to interpret and explain. We investigate which financial news topics are positively or negatively related to bank CDS spreads. Our findings indicate that especially negative news topics are significantly associated with changes in CDS spreads. These include not only financial aspects such as a decrease in dividends or revenue guidance but also non-financial topics such as sanctions and legal issues. Thereby, our understanding of which topics addressed in financial news are relevant for supporting decision-making in the context of credit risk management is improved.
This paper investigates the interrelationships and the asymmetric co-movements between the yield spread, macroeconomic factors and the stock market volatility across five major world economies. We highlight the non-linear adjusting process of the yield spread to its equilibrium value in response to changes in stock market volatility by using a consistent threshold cointegration error correction model. Our findings differ for different countries and for states of the economy. We find that for the US, the UK, Japan, and France, the adjustment of the yield spread towards its equilibrium value portrays the existence of negative asymmetric market volatility transmission. In addition, differences in the magnitude of the effects denote that yield spread changes in Japan and France appear to significantly adjust more swiftly to equilibrium values compared to the US where a higher degree of persistence is observed. Last, our results suggest evidence of bi-directional time varying Granger causality between the yield spread and stock market volatility for all countries, in both the pre- and post-crisis period.
News analytics measure the relevance, sentiment, novelty, and volume of news. They combine natural language analysis of content with historical and news metadata. Signals from analytics of this type have been shown to be predictive of volatility. Aggregation and filtering of news events can also generate alpha signals for portfolio management. Filters use thresholds set using both absolute and relative measures. This detects investor behavior associated with accumulation of information and changes in sentiment. The analytics described are used to generate investment signals. In practice, they would be combined with forecasts from other quantitative or research sources (e.g., factor, momentum, and earnings). In this chapter, we analyze investment signals derived only from news, an important distinction. Event studies on a broad universe of US equities (segmented by sector and capitalization class) are shown for the period 2003-2008. US portfolio simulation results are shown for these signals applied over 2006-2009. The portfolio simulation, like the event studies, is based on a "pure news" signal, without mixing in other quant signals, which can confuse the question of alpha from news. Both the event studies and portfolio simulation show evidence of exploitable alpha using news analytics.
In this paper, we investigate the relationship between stock prices and industrial production both for South and North of Euro-zone during the period 2004-2013. In contrast to previous studies we identify additional price interaction and dynamics investigating asymmetric adjustment behavior combined with long-run relationship using the Threshold cointegration approach. This method is proper as well because takes into consideration the type of shocks which appears in period 2004-2013. The results demonstrate symmetric adjustment process for the North and asymmetric for the South when stock prices and industrial production adjust to achieve the long-run equilibrium. The main cause of asymmetry is the difference in structural competitiveness which is weakest in South with respect to North. This finding is particularly important because provides the direction of economic policy that should adopt the governments of South of Euro-zone.
This study explores the forecasting performance of newswire messages, revealed by newspaper articles, for CDS. Five European countries with sovereign debt problems, daily data spanning the period 2009-2012, and ARIMA and ARIMAX modeling support the superiority of the ARIMAX model.
Time varying correlations are often estimated with multivariate generalized autoregressive conditional heteroskedasticity (GARCH) models that are linear in squares and cross products of the data. A new class of multivariate models called dynamic conditional correlation models is proposed. These have the flexibility of univariate GARCH models coupled with parsimonious parametric models for the correlations. They are not linear but can often be estimated very simply with univariate or two-step methods based on the likelihood function. It is shown that they perform well in a variety of situations and provide sensible empirical results.
We estimate the 'fundamental' component of euro area sovereign bond yield spreads, i.e. the part of bond spreads that can be justified by country-specific economic factors, euro area economic fundamentals, and international influences. The yield spread decomposition is achieved using a multi-market, no-arbitrage affine term structure model with a unique pricing kernel. More specifically, we use the canonical representation proposed by Joslin, Singleton, and Zhu (2011) and introduce next to standard spanned factors a set of unspanned macro factors, as in Joslin, Priebsch, and Singleton (2013). The model is applied to yield curve data from Belgium, France, Germany, Italy, and Spain over the period 2005-2013. Overall, our results show that economic fundamentals are the dominant drivers behind sovereign bond spreads. Nevertheless, shocks unrelated to the fundamental component of the spread have played an important role in the dynamics of bond spreads since the intensification of the sovereign debt crisis in the summer of 2011.
A review of news analytics and its applications in finance is given in this chapter. In particular we review the multiple facets of current research and some of the major applications. It is widely recognised news plays a key role in financial markets. The sources and volumes of news continue to grow. New technologies that enable automatic or semi-automatic news collection, extraction, aggregation and categorisation are emerging. Further machine learning techniques can be used to process the textual input of news stories to determine quantitative sentiment scores. We consider the various types of news available and how these can be processed to form inputs to financial models. We report applications of news, for prediction of abnormal returns, for trading strategies, for diagnostic applications as well as the use of news for risk control.
