2023
DOI: 10.1002/for.2944
|View full text |Cite
|
Sign up to set email alerts
|

Nowcasting inflation with Lasso‐regularized vector autoregressions and mixed frequency data

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… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 49 publications
0
3
0
Order By: Relevance
“…However, empirical applications of these proposals focus on nowcasting real GDP and not inflation. Recently, Aliaj et al (2023) have proposed a Lassoregularized MF-VAR (MF-VAR with machine learning method) that works reasonably well in nowcasting euro area HICP inflation. This is because the approach is based on a shrinkage method that can effectively handle large VARs by automatically excluding the information in the VAR that is deemed irrelevant.…”
Section: Mixed-frequency Vars (Mf-vars)mentioning
confidence: 99%
“…However, empirical applications of these proposals focus on nowcasting real GDP and not inflation. Recently, Aliaj et al (2023) have proposed a Lassoregularized MF-VAR (MF-VAR with machine learning method) that works reasonably well in nowcasting euro area HICP inflation. This is because the approach is based on a shrinkage method that can effectively handle large VARs by automatically excluding the information in the VAR that is deemed irrelevant.…”
Section: Mixed-frequency Vars (Mf-vars)mentioning
confidence: 99%
“…Typically, research in this field has focused on providing monthly GDP estimates; however, fuelled by the COVID-19 pandemic and the war in Ukraine, there has been increased interest in producing high-frequency measures of monthly data, such as inflation. Some studies in this field use traditional data sources such as weekly gasoline or commodity prices (Modugno, 2013;Breitung and Roling, 2015;Knotek II and Zaman, 2017;Clark, Leonard, Marcellino, and Wegmüller, 2022;Aliaj, Ciganovic, and Tancioni, 2023) and report robust forecasting and nowcasting gains compared to econometric benchmark models or market expectations. Another branch of the literature uses web-scraped price data to predict aggregate and disaggregate food price inflation (Macias, Stelmasiak, and Szafranek, 2023;Powell, Nason, Elliott, Mayhew, Davies, and Winton, 2018) and headline inflation (Harchaoui and Janssen, 2018;Aparicio and Bertolotto, 2020), again documenting improved forecasting accuracy.…”
Section: Literature Reviewmentioning
confidence: 99%
“…The energy component of the HICP consists of 14 price indices at the COICOP-10 level 8 that we try to match with the high-frequency price indicators discussed in the following. Most importantly, we use weekly price data from the Weekly Oil Bulletin (WOB) by the European Commission that has proven invaluable in previous work (Modugno, 2013;Aliaj et al, 2023). The WOB contains weekly information starting in 2005 about average fuel prices at the pump (Diesel, Supergrade petrol) and household-size deliveries of heating oil.…”
Section: Energy and Travel Servicesmentioning
confidence: 99%