Nowcasting inflation with Lasso‐regularized vector autoregressions and mixed frequency data
Оперативное прогнозирование инфляции с использованием векторных авторегрессий, регуляризованных методом Lasso, и данных смешанной частоты
2023-01-19
SCID: 54.1/rf22v2pr
Discuss with AI
Euro area inflationLasso-regularized vector autoregressionshigh-dimensional VARsinflation nowcastingmixed-frequency data
Figures from the paper
Abstract (AI)
Abstract We evaluate the predictive performances of the least absolute shrinkage and selection operator (Lasso) as an alternative shrinkage method for high‐dimensional vector autoregressions. The analysis extends the Lasso‐based multiple equations regularization to a mixed/high‐frequency data setting. Very short‐term forecasting (nowcasting) is used to target the Euro area's inflation rate. We show that this approach can outperform more standard nowcasting tools in the literature, producing nowcasts that closely follow actual data movements. The proposed tool can overcome information and policy decision problems related to the substantial publishing delays of macroeconomic aggregates.
Key Findings
1
Applied to Euro-area inflation, the approach produces very short-term nowcasts that closely track actual inflation movements.
2
It extends Lasso-based multiple-equation regularization to models using mixed-frequency and high-frequency data.
3
The proposed method can outperform standard nowcasting tools and mitigate information delays affecting macroeconomic monitoring and policy decisions.
4
The study evaluates Lasso regularization as an alternative shrinkage method for high-dimensional vector autoregressions.
Research Object
Euro area inflation rate
Research Subject
Very short-term predictive performance and nowcasting accuracy using Lasso-regularized vector autoregressions with mixed-frequency data
Publication Details
Publication Date
2023-01-19
Journal
Publisher
ISSN
Cited by
10
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest