Neural Forecasting of the Italian Sovereign Bond Market with Economic News
This work addresses bond market forecasting for investors or policymakers, but it is incremental as it applies existing neural methods to a specific dataset.
The paper tackled forecasting the Italian 10-year interest rate spread by incorporating economic news into a neural network, finding that a deep learning model based on LSTM outperformed classical machine learning methods and conventional determinants alone.
In this paper we employ economic news within a neural network framework to forecast the Italian 10-year interest rate spread. We use a big, open-source, database known as Global Database of Events, Language and Tone to extract topical and emotional news content linked to bond markets dynamics. We deploy such information within a probabilistic forecasting framework with autoregressive recurrent networks (DeepAR). Our findings suggest that a deep learning network based on Long-Short Term Memory cells outperforms classical machine learning techniques and provides a forecasting performance that is over and above that obtained by using conventional determinants of interest rates alone.