Associational mediation analysis has generally relied on the linearity of models to estimate the indirect effect as a product of regression coefficients. Very few examples of generalisations to nonlinear settings have been proposed, including a derivative-based method that, however, is far from being widely spread among scholars. In this paper, we clarify some aspects of such an approach to nonlinear mediation models, which have not been addressed by the previous literature. In addition, we run a simulation study to compare confidence intervals for the indirect effect obtained through different approaches.

Claudio Rubino, & Chiara Di Maria (2022). Insights into the derivative-based method for nonlinear mediation models. In Book of Short Papers - SIS2022.

Insights into the derivative-based method for nonlinear mediation models

Claudio Rubino;Chiara Di Maria
2022

Abstract

Associational mediation analysis has generally relied on the linearity of models to estimate the indirect effect as a product of regression coefficients. Very few examples of generalisations to nonlinear settings have been proposed, including a derivative-based method that, however, is far from being widely spread among scholars. In this paper, we clarify some aspects of such an approach to nonlinear mediation models, which have not been addressed by the previous literature. In addition, we run a simulation study to compare confidence intervals for the indirect effect obtained through different approaches.
Settore SECS-S/01 - Statistica
Claudio Rubino, & Chiara Di Maria (2022). Insights into the derivative-based method for nonlinear mediation models. In Book of Short Papers - SIS2022.
File in questo prodotto:
File Dimensione Formato  
insights_manuscript sis 2022.pdf

non disponibili

Descrizione: Manoscritto pubblicato
Tipologia: Versione Editoriale
Dimensione 162.11 kB
Formato Adobe PDF
162.11 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/10447/566243
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact