Search

“Structural Equation Modeling is one of multivariate statistical method that used to explain multiple relationships between latent variables simultaneously to test a mediation model to conduct a formal test on mediation effects. Application PLS-SEM for exploratory research and theory development are increasing. Under certain conditions, the effect of exogenous variables on endogenous variable is also strengthened or weakened by moderating variable. In SEM, there are two approaches in analyzing moderation variables, namely the interaction method and the multigroup method. This article aims to compare the mediation effect on interaction approaches and multigroup approaches in Structural Equation Modeling. The data used is the case of timeliness of Bank X mortgage payments. In this article, statistical methods are evaluated to compare indirect effect between groups and examine indirect effect on each group. It was concluded that Collectability Status moderates the indirect relationship between Capital and the Timeliness of Payment through Willingness to Pay. Debtors with current collectability status more strongly effect the Timeliness of Payment than debtors with incurrect collectability status. Theresults of testing indirect effects on moderation with interaction and multigroup approaches are not much different. In the multigroup approach, the bootstrap interval bias is smaller than the bootstrap interval bias in the interaction approach. The Q-square Predictive Relevance value in both methods is quite high, indicating that the model is good. On the Current Collectibility Status group Q^2 is 89.3%, in the incurrect Collectibility Status Q^2 is 84.2%. While in the interaction approach, Q^2 is 70.4%. Researcher recommend a multigroup approach to data that has categorical moderation variables because differences between groups can be directly observed without adding interaction variables in the model.”

Geographically Weighted Panel Regression (GWPR), a combination of panel regression and geographically weighted regression (GWR), is used to analyze panel data and capture diverse relationship between locations. GWPR was developed on data with panel-fixed effects and applied to modeling data with spatial heterogeneity and time series. One method for estimating parameters in the GWPR is weighted least squares (WLS), which are sensitive to outliers. The present study aimed to use the M- method to estimate GWPR model parameters in data containing outliers using fixed-effect GWPR modeling for the sugar cane yield in East Java of Indonesia from 2019 to 2021. Sugarcane yield data in East Java contained outliers in several areas, including Malang, Blitar, and Ngawi Districts. Because the data contains outliers, a robust method with the M estimator was applied. The results showed that plantation areas significantly affected production in all districts.The R2 of the model was 0.87, showing that GWPR model with M estimation was appropriate in predicting sugarcane yield. Based on the Akaike Information Criterion (AIC) value, the GWPR model with M estimation had better performance than GWPR model alone.

Semiparametric path analysis is a combination of parametric and nonparametric path analysis. Semiparametric path analysis is used when there are partially nonlinear and unknown patterns of relationships. One approach to semiparametric pathways is truncated spline. Truncated spline approach tends to search for their own estimation of regression functions according to the data. This is because in the truncated spline there are knot points, which are intersection points that indicate changes in data behavior patterns. Truncated spline semiparametric path analysis will be applied to this study to determine the variables that have a significant effect on the development of the Cashless Society so that the result can be used as a reference for banks and the government in maximizing non-cash-based community development. The data used is the result of a questionnaire with 100 respondents of mobile banking users in Jakarta and will be analyzed using R Studio. Based on the results, it was found that the optimal knot point in the truncated spline function is 3 with many knots is 1, thus dividing the condition of digitizing electronic money into 2 regimes. It was concluded that the product and digitalization of electronic money had a significant effect on the development of cashless society where the modeling obtained could explain 83.87548% of the data. However, when electronic digitalization increases through the value of knot points, the development of cashless society tends to stagnate. This could be due to people who are not ready when the condition of digitizing electronic money is increasingly sophisticated because the available electronic money features are increasingly complex. Therefore, it is important for banks to pay attention to the sophistication of electronic money features provided to customers and adjust the target market so that customers are more accustomed and comfortable to use electronic money in the future.