%Gra: an SAS macro for generalized redundancy analysis. Issue 5 (24th March 2017)

• Record Type: Journal Article
• Authors: Lovaglio, Pietro Giorgio; Vacca, Gianmarco
• Is Part Of: Journal of statistical computation and simulation. Volume 87:Issue 5(2017)
• Page Start: 1048
• Page End: 1060
• Publication Date: 2017-03-24
• Subjects: Redundancy analysis -- SAS macro -- alternating least squares -- latent components -- reduced rank regression; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5028505
• DOI: 10.1080/00949655.2016.1243685
• Languages: English
• ISSNs: 0094-9655

Abstract

ABSTRACT: In the framework of redundancy analysis and reduced rank regression, the extended redundancy analysis model managed to account for more than two blocks of manifest variables in its specification. A further extension, the generalized redundancy analysis (GRA), has been recently proposed in literature, with the aim of incorporating external covariates into the model, thanks to a new estimation algorithm that manages to separate all the contributions of the exogenous and external covariates in the formation of the latent composites. At present, software to estimate GRA models is not available. In this paper, we provide an SAS macro, %GRA, to specify and fit structural relationships, with an application to illustrate the use of the macro.