Abstract In this paper, we generalize the Extended Redundancy Analysis (ERA), extending this new class of models to allow for external covariate effects. In particular,covariates are allowed to affect endogenous indicators indirectly through the composites and/or directly. The method proposed herein is called Generalized Redundancy Analysis (GRA), which allows us to specify and fit a variety of relationships among composites and endogenous variables. To illustrate the advantages of GRA over ERA we propose two simulation studies. Other than the proposal of GRA, a second aspect of originality of this paper is that, to our knowledge, no existing empirical research addresses the behaviour of ERA with external covariate effect in simulation studies.
Lovaglio, P., Boselli, R. (2013). Generalized Redundancy Analysis. In E. Brentari, M. Carpita (a cura di), Advances in Latent Variables (pp. 1-8). Vita e Pensiero.
Generalized Redundancy Analysis
LOVAGLIO, PIETRO GIORGIO;BOSELLI, ROBERTO
2013
Abstract
Abstract In this paper, we generalize the Extended Redundancy Analysis (ERA), extending this new class of models to allow for external covariate effects. In particular,covariates are allowed to affect endogenous indicators indirectly through the composites and/or directly. The method proposed herein is called Generalized Redundancy Analysis (GRA), which allows us to specify and fit a variety of relationships among composites and endogenous variables. To illustrate the advantages of GRA over ERA we propose two simulation studies. Other than the proposal of GRA, a second aspect of originality of this paper is that, to our knowledge, no existing empirical research addresses the behaviour of ERA with external covariate effect in simulation studies.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
