No abstract

No abstract

We test the usefulness of machine learning (ML) for the valuation and pricing of sovereign risk in the Euro area along two important dimensions: i) its predictive accuracy compared with traditional econometric methods, and ii) its assessment of the main economic factors underlying market perceptions of sovereign risk.We find that ML techniques can capture the dynamics inherent in the market valuation of country risk far more efficiently than traditional econometric models, both in the cross-section and in the time series. Moreover, we show that public sentiment about financial news, redenomination fears and the degree of hawkishness/dovishness expressed in the ECB president's speeches are major contributors to sovereign bond spreads. We also confirm that macroeconomic and global financial factors affect sovereign risk assessment and the corresponding formation of sovereign spreads.

No abstract