Directed graphs and variable selection in large vector autoregressive models

Dominik Bertsche, Ralf Brüggemann and Christian Kascha

Journal of Time Series Analysis, 2023, vol. 44, issue 2, 223-246

Abstract: We represent the dynamic relation among variables in vector autoregressive (VAR) models as directed graphs. Based on these graphs, we identify so‐called strongly connected components. Using this graphical representation, we consider the problem of variable choice. We use the relations among the strongly connected components to select variables that need to be included in a VAR if interest is in impulse response analysis of a given set of variables. Our theoretical contributions show that the set of selected variables from the graphical method coincides with the set of variables that is multi‐step causal for the variables of interest by relating the paths in the graph to the coefficients of the ‘direct’ VAR representation. An empirical application illustrates the usefulness of the suggested approach: Including the selected variables into a small US monetary VAR is useful for impulse response analysis as it avoids the well‐known ‘price‐puzzle’.

Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1111/jtsa.12664

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:bla:jtsera:v:44:y:2023:i:2:p:223-246

Ordering information: This journal article can be ordered from
http://www.blackwell ... bs.asp?ref=0143-9782

Access Statistics for this article

Journal of Time Series Analysis is currently edited by M.B. Priestley

More articles in Journal of Time Series Analysis from Wiley Blackwell
Bibliographic data for series maintained by Wiley Content Delivery ().