Predictive Power of Composite Socioeconomic Indices in Regression and Classification: Principal Components and Partial Least Squares

Stefanía D’Iorio, Liliana Forzani, Rodrigo García Arancibia and Ignacio Girela
Additional contact information
Stefanía D’Iorio: Universidad Nacional de Entre Ríos
Liliana Forzani: Universidad Nacional del Litoral/ CONICET
Ignacio Girela: Universidad Nacional de Córdoba/ CONICET

No 246, Working Papers from Red Nacional de Investigadores en Economía (RedNIE)

Abstract: Principal Components Analysis (PCA) and Partial Least Squares (PLS) have been used for the construction of socioeconomic status (SES) indices to use as a predictor of the well-being status in targeted programs. Generally,these indicators are constructed as a linear combination of the first component. Due to the characteristics of the socioeconomic data, different extensions of PCA and PLS for non-metric variables have been proposed for these applications. In this paper we compare the predictive performance of SES indices constructed using more than one component. Additionally, for the inclusion of non-metric variables, a variant of the normal mean coding is proposed that takes into account the multivariate nature of the variables, that we call multivariate normal mean coding (MNMC). Using simulations and real data, we found that PLS using MNMC as well as the classical dummy encoding method give the best predictive results with a more parsimonious SES index.

Keywords: Dimension Reduction; Categorical Predictors; SES; Proxy Mean Test (search for similar items in EconPapers)
Pages: 31 pages
Date: 2023-05
New Economics Papers: this item is included in nep-ets
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://rednie.eco.unc.edu.ar/files/DT/246.pdf (application/pdf)

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:aoz:wpaper:246

Access Statistics for this paper

More papers in Working Papers from Red Nacional de Investigadores en Economía (RedNIE) Contact information at EDIRC.
Bibliographic data for series maintained by Laura Inés D Amato ().