Modeling indices using partial least squares: How to determine the optimum weights?

Creative Commons License

DİRSEHAN T., Henseler J.

Quality and Quantity, vol.57, pp.521-535, 2023 (Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 57
  • Publication Date: 2023
  • Doi Number: 10.1007/s11135-022-01515-5
  • Journal Name: Quality and Quantity
  • Journal Indexes: Scopus, International Bibliography of Social Sciences, ABI/INFORM, Index Islamicus, Political Science Complete, Psycinfo, Social services abstracts, Sociological abstracts, Worldwide Political Science Abstracts
  • Page Numbers: pp.521-535
  • Keywords: Composite measurement, Indices, Partial least squares path modeling, PLS Mode A, PLS Mode B, Weighting schemes
  • Open Archive Collection: AVESIS Open Access Collection
  • Marmara University Affiliated: Yes

Abstract

© 2022, The Author(s).Indices are often used to model theoretical concepts in economics and finance. Beyond the econometric models used to test the relationships between these variables, partial least squares path modeling (PLS-PM) allows the study of complex models, but it is an estimator that is still in its infancy in economics and finance research. Thus, the use of PLS-PM for composite analysis needs to be explored further. As one such attempt, this paper is focused on the determination of the indices’ optimum weights. For this purpose, the effects of the market potential index (MPI) on foreign direct investment (FDI) and gross domestic product (GDP) were analysed by implementing different weighting schemes. The assessment of the model shows that PLS Mode B leads to better model fit.