Segnalazione dalla newsletter BCE-ECB

(This is) the latest issue (No. 112) of the ECB Research Bulletin, entitled “Forecasting euro area inflation with machine learning models”.

Michele Lenza (Directorate General Research, European Central Bank), Inés Moutachaker (INSEE) and Joan Paredes (Directorate General Economics, European Central Bank) show how a new machine learning model for inflation forecasting, recently developed at the ECB, captures very general forms of non-linearity, such as changes in the sensitivity of inflation dynamics to the prevailing economic circumstances. 

Forecasts from this machine learning model closely track Eurosystem staff inflation projections, suggesting that these projections capture mild non-linearity in inflation dynamics – likely owing to expert judgement – and are in line with state-of-the-art econometric methodologies.

The ECB Research Bulletin features a selection of recent work on policy-relevant topics by ECB economists. Published on a monthly basis, the articles in the Research Bulletin are intended for a general audience. The views expressed in each article are those of the authors and do not necessarily represent the views of the European Central Bank and the Eurosystem. (...)