A Marginal Maximum Likelihood Approach for Extended Quadratic Structural Equation Modeling with Ordinal Data

Abstract

The literature on non-linear structural equation modeling is plentiful. Despite this fact, few studies consider interactions between exogenous and endogenous latent variables. Further, it is well known that treating ordinal data as continuous produces bias, a problem which is enhanced when non-linear relationships between latent variables are incorporated. A marginal maximum likelihood-based approach is proposed in order to fit a non-linear structural equation model including interactions between exogenous and endogenous latent variables in the presence of ordinal data. In this approach, the exact gradient of the approximated observed log-likelihood is calculated in order to attain the approximated maximum likelihood estimator. A simulation study shows that the proposed method provides estimates with low bias and accurate coverage probabilities.

The impact of sustainable development strategy on sustainable supply chain firm performance in the digital transformation era

“…This work uses structural equation modeling (SEM) to study SCC, SDS, DIT, and COA's combined effect on sustainable SC firm performance. SEM is an appropriate approach for studying interrelationships between independent and dependent variables as the method involves regression analysis (Dubey et al, 2020; Jin et al, 2020). Additionally, SEM is suitable when one regression analysis can also be an independent variable for another.…”

Analyzing the mediating role of organizational ambidexterity and digital business transformation on industry 4.0 capabilities and sustainable supply chain performance

“…According to Fornell and Larcker (1981), SEM is an appropriate approach, as the study requires conducting a set of regression analyzes. SEM includes statistical procedures for measurement testing, functional, predictive and causal hypotheses (Jin et al , 2020). Furthermore, SEM enables the calculation of measurement error variance in both exogenous and endogenous variables, which fits this study’s need.…”

Artificial intelligence-driven innovation for enhancing supply chain resilience and performance under the effect of supply chain dynamism: an empirical investigation

“…Structural equation modeling (SEM) includes statistical procedures for measurement testing, functional, predictive, and causal hypotheses (Jin et al 2020 ). SEM is a suitable approach for examining interrelationships between one or more independent variables and one or more dependent variables since the method includes a set of regression analyses (Jin et al 2020 ; Dubey et al 2020 ). Furthermore, SEM is appropriate when the dependent variable for one regression analysis could become an independent variable of another.…”

“…Besides, SEM allows the computation of measurement error variance in both exogenous and endogenous variables, which meets this study's need. Therefore, we proposed to use SEM to test our hypothesis using Maximum Likelihood Estimation (MLE) (Cudeck 1989 ) as recommended by Jin et al ( 2020 ). All the SEM computations were performed on IBM SPSS ver.…”