Data Preselection in Machine Learning Methods: An Application to Macroeconomic Nowcasting with Google Search Data

Agostino Capponi, Charles-Albert Lehalle, Anna Simoni () and Laurent Ferrara
Additional contact information
Anna Simoni: CNRS - Centre National de la Recherche Scientifique

Post-Print from HAL

Abstract: In this chapter, we present some Machine Learning (ML) econometric methods that allows to conveniently exploit Google search data for macroeconomic nowcast. In particular, we focus on the issue of variables preselection among a large set of Google search categories before entering them into ML approaches in order to nowcast macroeconomic variables. We consider two ML approaches allowing to estimate linear regression models starting from large information sets: a factor extraction and a Ridge regularisation. As an application we consider euro area GDP growth nowcasting using weekly Google Search data. Empirical results tend to suggest that estimating a Ridge regression associated with an ex ante preselection procedure appears as a pertinent strategy in terms of nowcasting accuracy.

Date: 2023-04-30
References: Add references at CitEc
Citations:

Published in Machine Learning and Data Sciences for Financial Markets, 1, Cambridge University Press, pp.490-506, 2023, ⟨10.1017/9781009028943.026⟩

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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:hal:journl:hal-04369062

DOI: 10.1017/9781009028943.026

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().