New study on model specification search in GSCA
16.05.2022
We are happy to announce the publication of our paper titled “A Prediction-Oriented Specification Search Algorithm for Generalized Structured Component Analysis” in Structural Equation Modeling: A Multidisciplinary Journal. In this paper, Gyeongcheol Cho (McGill University), Heungsun Hwang (McGill University), Marko Sarstedt (LMU Munich), and Christian M. Ringle (Hamburg University of Technology) present a new approach for model specification search for generalized structured component analysis (GSCA) that explores potential model relationships from a prediction perspective. Their predictive feedforward algorithm automatically searches for the best combination of predictors that minimizes the target variables’ expected prediction error. The authors test the algorithm’s efficacy using simulated data and illustrate its performance using a standard technology acceptance model from information systems.
Access at: https://www.tandfonline.com/doi/full/10.1080/10705511.2022.2057315