We nowcast world trade using machine learning, distinguishing between tree-based methods (random forest, gradient boosting) and their regression-based counterparts (macroeconomic random forest, linear gradient boosting). While much less used in the literature, the latter are found to outperform...

We nowcast world trade using machine learning, distinguishing between tree-based methods (random forest, gradient boosting) and their regression-based counterparts (macroeconomic random forest, linear gradient boosting). While much less used in the literature, the latter are found to outperform...

We propose a new methodology to recover firm-time varying financial constraints from firms' production behavior. We model financial constraints as the profitability that firms forgo when budget constraints on production inputs bind, impeding them from using the optimal level of inputs and...

We propose a new methodology to recover firm-time varying financial constraints from firms' production behavior. We model financial constraints as the profitability that firms forgo when budget constraints on production inputs bind, impeding them from using the optimal level of inputs and...

Until now, stock market responses to a distress scenario for oil prices have been analysed considering prices in domestic currency. This assumption implies merging the commodity risk with the exchange rate risk when oil and stocks are traded in different currencies. This article proposes...

Macroeconomic imbalances increase the vulnerability of an economy to adverse shocks, which in turn can lead to crises with severe economic and social costs. We propose an early warning model that predicts such crises. We identify a set of macroeconomic indicators capturing domestic and external...

This paper examines which measures of financial conditions are informative about the tail risks to output growth in the euro area. The Composite Indicator of Systemic Stress (CISS) is more informative than indicators focusing on narrower segments of financial markets or their simple aggregation...