Data
Research Blog Events
About Careers Newsroom Get Started

News Sentiment Analysis for Fixed Income Markets

In this report, we summarize eight academic papers to illustrate the ways in which we can use news sentiment analysis for a variety of fixed income markets applications.

The first three papers we highlight take similar approaches when incorporating macroeconomic news sentiment into corporate and sovereign credit valuation and risk analysis. One paper approaches CDS spread valuation by using news sentiment, while another looks at cross-asset network effects of news propagation. Another paper looks at how aggregate corporate earnings can assist with municipal bond credit valuation.

1. Enhanced Corporate Bond Yield Modelling incorporating Macroeconomic News Sentiment

The study investigates the predictive properties of news sentiment indicators on European corporate bond yield spreads by using the RavenPack News Analytics sentiment data. It examines 10 bonds from 7 different companies covering the period from January 2007 to May 2017. The results from the study show that including news sentiment indicators as predictive variables improves yield spread forecast performance.

2. Macroeconomic News Sentiment: Enhanced Risk Assessment for Sovereign Bonds

In a similar fashion, this study enhances the modelling and risk assessment of sovereign bond spreads by taking into account quantitative information gained exclusively from daily macro-economic news sentiment. The authors investigate sovereign bond spreads of over 300 bonds in five European countries from 2007 to 2017 and improve the prediction of spread changes by incorporating news sentiment from relevant entities and macro-economic topics.

3. Forecasting sovereign bond spreads with macroeconomic news sentiment

This is an earlier paper done by the same team as above, and applies macroeconomic news sentiment-based prediction on sovereign bond spread with a specific focus on German sovereign bonds. Similarly, they create sentiment time series to mirror the mood in the news regarding political and economic issues in European countries and analyse positive and negative sentiment separately, taking into account market restrictions and trading venues.

4. News Sentiment and Credit Risk Valuation: Evidence from the CDS Market

The study takes a similar, but a rather more granular approach, to CDS spreads. CDS contracts are typically traded with at least $10 million in notional value, and exclusively by institutional investors, which allows for testing the relation between news sentiment and institutional credit investor reactions. CDS spreads represent a cleaner measure of credit risk, as they are standardised and less subject to liquidity premium relative to individual bond issues.

Using RavenPack’s News Analytics data spanning the period from 2001 to 2016, this study examines and demonstrates a strong negative relationship between news sentiment and CDS spreads.

5. What Economic Factors Underlie Connectedness in Corporate Credit Default Swaps: News vs. Macroeconomic Factors?

This study is a bit more theoretical in nature, rather than immediately practical, but nevertheless useful in its efforts to illustrate the complex ways in which news and events propagate across networks of assets. The authors use network theory to examine the economic factors that underlie return and price volatility networks between 89 corporate CDS contracts over the period of 2007 to 2013. They use company news measures and macroeconomic covariates to explore how their effects propagate across the network of assets and affect overall net connectedness.

6. The Information Externality of Corporate Financial Information in the Secondary State-Bond Market

This study tackles the informational opaqueness problem that characterises the municipal state-bond market and provides evidence on the role of corporate financial information in predicting muni bond returns. In particular, the author investigates the informational role of state-wide aggregate earnings and finds that corporate-earnings changes that are aggregated at the state level are leading indicators of regional economic conditions and muni bond returns, and are generally timelier than other economic signals.

These findings suggest that statewide aggregate earnings provide bondholders with real-time signals about regional economic performance and find the effects to be stronger when corporate managers disseminate earnings news more extensively.

7. The Effect of Credit Rating Changes on Voluntary Disclosure

In this study, the authors provide evidence suggesting that corporate credit rating changes influence firms’ voluntary disclosure behavior. While other studies, unsurprisingly, show that quantitative and qualitative firm disclosures affect credit ratings, they document a negative relation between the direction of the credit rating change and the provision of voluntary disclosure in both directions - firms respond to exogenous downgrades by increasing voluntary disclosure and to exogenous upgrades by decreasing voluntary disclosure. Their analyses suggest that the mechanism underlying this negative relation is the financing frictions that arise from credit rating changes.

8. CDS Trading and Relationship Lending Dynamics

This study investigates how the development of the credit default swap (CDS) market affects lenders’ incentives to initiate new lending relationships. The study’s findings also shed light on how CDS trading affects the evolution of lending relationships.

Access The Full Report on Sentiment Analysis for Fixed Income Markets

Access the full report “News Sentiment Analysis for Fixed Income Markets: the Top 8 Academic Papers” here.

Request White Paper

Request White Paper

We will process your personal data with the purpose of managing your personal account on RavenPack and offering our services. You can exercise your rights of access, rectification, erasure, restriction of processing, data portability and objection by emailing us at privacy@ravenpack.com. For more information, you can check out our Privacy Policy.

Discover the RavenPack Platform

Choose from over 190 curated datasets covering a variety of different topics.

Explore Now