Publication:
Econometric methods for business cycle dating

Thumbnail Image
Files
BCDating.pdf(836.4 KB)
Date
2023-12-13
relationships.isAuthorOfPublication
relationships.isSecondaryAuthorOf
relationships.isDirectorOf
Authors
Camacho, Maximo ; Gadea, M. Dolores
item.page.secondaryauthor
item.page.director
Publisher
publication.page.editor
publication.page.department
Métodos Cuantitativos para la Economía y la Empresa
DOI
https://doi.org/10.1093/acrefore/9780190625979.013.885
item.page.type
info:eu-repo/semantics/article
Description
© 2023. This document is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ This document is the submitted version of a published work that appeared in final form in Oxford Research Encyclopedia of Economics and Finance.
Abstract
Business cycle dating helps in developing economic analysis and is useful for economic agents whether they be policy makers, investors or academics. This paper reviews old and recent research on dating the reference cycle turning points and is intended as a guide to the applied researcher. All these methods provide a statistical alternative to cycle dating committees, although full automatism and researcher’s art could be complements rather than substitutes in some dating scenarios. Our survey divides the dating literature into two groups with different approaches to dating the business cycle from a set of coincident economic indicators: averagethen- date or date-then average. In both cases, the dating techniques can be divided into nonparametric and parametric. The paper shows the theoretical foundations of both types of techniques and describes in detail the algorithms or estimation methods necessary for their implementation. Finally, the paper describes empirical applications of the different methods with data of different frequencies, trying to show how they work in practice and pointing out their advantages and disadvantages. This empirical illustrations include a compilation of the codes in different languages (R, Matlab or Gauss). In our opinion, future research should focus on developing methods that are robust to changes in volatility or large outliers and on exploring the usefulness of big data sources and the classification ability offered by machine learning methods.
publication.page.subject
Business cycles , Dating methods , Classification techniques , Non-parametric , Parametric
Citation
Oxford Research Encyclopedia of Economics and Finance
URI
http://hdl.handle.net/10201/137230
item.page.embargo
Collections
Artículos
Ir a Estadísticas