Proposed Method to Employ the Robust Correlation in Structural Equations Modeling

Main Article Content

Saif Ramzi Ahmed, Bashar A. Al-Talib

Abstract

The idea of ​​the research is to fortify the path coefficients in the structural equations model through the use of the robust correlation coefficients. Three methods of the robust correlation were compared in addition to the Pearson correlation. A comparison has been made between the proposed method and the traditional method through a causal model proposed by the researchers. Several simulation scenarios have been applied in the experimental aspect. In order to test the efficiency of the generated models, samples of different sizes (n=20,100,1000) were used. Furthermore, the explanatory variables in the causal model were contaminated with different contamination percentages. Path parameters were estimated using the four-fold correlation matrix and the results were compared through some statistical criteria. The study concluded that the pathway coefficients method using the Robust Correlation has given the best results compared to the traditional methods.

Article Details

Section
Articles