Journal of Statistics Applications & Probability


This study aims to model the path analysis of linear, quadratic, and cubic truncated splines at one and two knot points using Iteratively Reweighted Least Square (IRLS) and determine the best model in the case of Community Welfare with the Role of the Government as a mediating variable. The data used in this study is primary data. Data analysis used in this study is nonparametric path analysis with Iteratively Reweighted Least Square (IRLS) using Rstudio software. The estimation of the nonparametric path model function truncated spline of linear, quadratic, and cubic 1 and 2 point knots indicates that changes in data behavior patterns occur when the character / personality of the debtor is quite good and good at paying credit on time. The novelty of this research is to model the development of nonparametric path analysis, namely nonparametric truncated spline analysis of linear, quadratic, and cubic orders using Iterative Reweighted Least Square (IRLS) on community welfare data.

Digital Object Identifier (DOI)