Brazilian Business Cycle Analysis in a High-Dimensional and Time-Irregular Span Context

André Nunes Maranhão ()
Additional contact information
André Nunes Maranhão: Getulio Vargas Foundation São Paulo School of Economics

Journal of Business Cycle Research, 2024, vol. 20, issue 1, No 1, 58 pages

Abstract: Abstract A recent challenge in business cycle analysis is high-dimensional databases with time-irregular span of available time-series starting points (TISN). Using the Brazilian business cycle, this study proposes a hyper-dimension environment entropic test of relative information, and develops a time-irregular span version for the generalized dynamic factor model (GDFM) of Forni et al. (Rev Econ Stat, 82(4):540–554, 2000). We estimate the GDFM model and its version for time-irregular series (TISN-GDFM) using 2571 series from January 1980 to December 2017. The results show that the TISN-GDFM model has gained commonality over time, better describing the Brazilian business cycle. The results show a larger set of time series selected in the current quarter than in the other lags, concentrated in categories related to Confidence and Surveys, Credit, Industry, Sectoral, and Financial series. We categorize the series for a better description of the Brazilian business cycle. The estimated Brazilian business cycle with TISN-GDFM shows evidence of high commonality for the retail category, being pro-cyclical, and leading. Analyzing the disaggregated series, the pro-cyclical and antecedent behavior of credit default and delays are highlighted. Considering only the chronologically complete series, we estimate the GDFM model. The dimension reduction test results show the use of a large portion of all available series. The estimated Brazilian business cycle has had Climate and Industry categories with the highest commonality. As new series are incorporated into the TISN-GDFM model, better adjustment of the estimated cycle is verified. When incorporating new series, there is a significant improvement in the description of the Brazilian business cycle.

Keywords: Brazilian business cycle; Hyper-dimensional entropic relative information test; Generalized dynamic factor model; Time-irregular span (search for similar items in EconPapers)
JEL-codes: C14 C55 E32 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s41549-024-00095-7 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:spr:jbuscr:v:20:y:2024:i:1:d:10.1007_s41549-024-00095-7

Ordering information: This journal article can be ordered from
http://www.springer. ... nomics/journal/41549

DOI: 10.1007/s41549-024-00095-7

Access Statistics for this article

Journal of Business Cycle Research is currently edited by Michael Graff

More articles in Journal of Business Cycle Research from Springer, Centre for International Research on Economic Tendency Surveys (CIRET)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().