Presented By O’Reilly and Cloudera
Make Data Work
21–22 May 2018: Training
22–24 May 2018: Tutorials & Conference
London, UK

Macroeconomic news sentiment: Enhanced risk assessment for sovereign bond spreads

Christina Erlwein-Sayer (OptiRisk Systems)
Data-driven business management
Location: Capital Suite 4
Average rating: ***..
(3.62, 8 ratings)

Christina Erlwein-Sayer explains how to enhance the modeling and forecasting of sovereign bond spreads by considering quantitative information gained from macroeconomic news sentiment, using a number of large news analytics datasets. Christina then offers a demonstration of this model that investigates sovereign bonds spreads of five European countries.

The prediction of spread changes is improved by incorporating news sentiment from relevant entities and macroeconomic topics. In particular, daily news sentiment series are created from sentiment scores as well as positive and negative news volume, and their effects on yield spreads and spread volatility are investigated. Christina conducts a rolling correlation analysis between sovereign bond spreads and accumulated sentiment series and analyzes changing correlation patterns over time, detecting market regimes through correlation series and exploring the impact of news sentiment on sovereign bonds in different market circumstances. Along the way, Christina shares best-suited external news variables for forecasts in an ARIMAX model setup. Error measures for forecasts of spread changes and volatility proxies are improved when sentiment is considered. These findings are then utilized to monitor sovereign bonds from European countries and detect changing risks through time.

Photo of Christina Erlwein-Sayer

Christina Erlwein-Sayer

OptiRisk Systems

Christina Erlwein-Sayer is senior quantitative analyst and researcher at OptiRisk Systems, where she works on financial analytics, in particular, models and tools for portfolio construction and fixed income risk assessment. Her main research interest lies in regime-switching models for finance. Previously, she was a researcher and consultant in the Financial Mathematics Department at Fraunhofer ITWM, Kaiserslautern, Germany. She currently leads the Eurostars project SenRisk on the development of sentiment-enhanced risk assessment tools. She is also member of Quantess London, the first community in London designed for female quants who have a passion in data, science, and finance. Quantess mentors and promotes females from across seniority and industries with a passion in quantitative fields to discuss practical research ideas. Christina holds a PhD in mathematics from Brunel University, London.