Forecasting euro area inflation with machine-learning models

Michele Lenza, Inès Moutachaker and Joan Paredes

Research Bulletin, 2023, vol. 112

Abstract: Inflation forecasts and their risks are key for monetary policy decisions. The strategy review concluded in 2021 highlighted how most Eurosystem models used to forecast inflation are linear. Linear models assume that changes in, for example, wages, always have the same fixed, proportional effect on inflation. A new machine learning model, recently developed at the ECB, captures very general forms of non-linearity, such as a changing sensitivity of inflation dynamics to 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. JEL Classification: C52, C53, E31, E37

Keywords: Inflation; Non-linearity; Quantile Regression Forest (search for similar items in EconPapers)
Date: 2023-10
Note: 411196
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.ecb.europa.eu//pub/economic-research/r ... 7~b910853393.en.html (text/html)
https://www.ecb.europa.eu//pub/economic-research/r ... 17~b910853393.en.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ecb:ecbrbu:2023:0112:

Access Statistics for this article

More articles in Research Bulletin from European Central Bank 60640 Frankfurt am Main, Germany. Contact information at EDIRC.
Bibliographic data for series maintained by Official Publications ().