ExtremeBounds: Extreme Bounds Analysis in R

Marek Hlavac

Journal of Statistical Software, 2016, vol. 072, issue i09

Abstract: This article introduces the R package ExtremeBounds to perform extreme bounds analysis (EBA), a sensitivity test that examines how robustly the dependent variable of a regression model is related to a variety of possible determinants. ExtremeBounds supports Leamer's EBA that focuses on the upper and lower extreme bounds of regression coefficients, as well as Sala-i-Martin's EBA which considers their entire distribution. In contrast to existing alternatives, it can estimate models of a variety of user-defined sizes, use regression models other than ordinary least squares, incorporate non-linearities in the model specification, and apply custom weights and standard errors. To alleviate concerns about the multicollinearity and conceptual overlap of examined variables, ExtremeBounds allows users to specify sets of mutually exclusive variables, and can restrict the analysis to coefficients from regression models that yield a variance inflation factor within a prespecified limit.

Date: 2016-08-30
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (24)

Downloads: (external link)
https://www.jstatsoft.org/index.php/jss/article/view/v072i09/v72i09.pdf
https://www.jstatsoft.org/index.php/jss/article/do ... ounds_0.1.5.2.tar.gz
https://www.jstatsoft.org/index.php/jss/article/do ... ile/v072i09/v72i09.R

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:jss:jstsof:v:072:i09

DOI: 10.18637/jss.v072.i09

Access Statistics for this article

Journal of Statistical Software is currently edited by Bettina Grün, Edzer Pebesma and Achim Zeileis

More articles in Journal of Statistical Software from Foundation for Open Access Statistics
Bibliographic data for series maintained by Christopher F. Baum ().