Indicadores Coincidentes para a Atividade Industrial Brasileira Baseado em Modelos Vetoriais Auto-Regressivos de Freqüência Mista: comparação de metodologias

Gilberto Hollauer and João Issler

No 1198, Discussion Papers from Instituto de Pesquisa Econômica Aplicada - IPEA

Abstract: Very useful information, usually ignored, for construction of coincident index is the target quarterly series itself. This can be very inefficient because typically the monthly coincident series keep just high economical correlation, not always tested, with the quarterly target series. Actually, the construction of a mixed-frequency coincident index is statistically complicated. On the ground that, Mariano e Murasawa presented a new methodology, in which this information is aggregated, and the quarterly variable is assumed as a latent variable observed just every three months. In this fashion, is implicitly assumed that the monthly GDP not observed represents the state of industrial economy. Usually a common factor is imposed, as in the Stock-Watson approach. Another possible approach more attractive is not to impose any common factor and instead, like Mariano e Murosawa, to build a mixed frequency VAR model without common factor. In this work we compare the performance in-sample of the mixed frequency VAR model for building coincident indicator with others approaches, dating the business cycle for industrial activity with Bry-Boschan procedure.

Pages: 29 pages
Date: 2006-07
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.ipea.gov.br/portal/images/stories/PDFs/TDs/td_1198.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:ipe:ipetds:1198

Access Statistics for this paper

More papers in Discussion Papers from Instituto de Pesquisa Econômica Aplicada - IPEA Contact information at EDIRC.
Bibliographic data for series maintained by Fabio Schiavinatto ().