One of the benchmarks to see the quality of education and human resources in Indonesia is the average length of schooling. If the school average is higher, it can positively impact Indonesian society, enabling it to compete globally. There are several factors, both economic and educational factors, that influence the low average length of schooling in Central Java Province. Therefore, this research aims to model and determine what variables can influence the average length of schooling in Central Java in 2023 using a nonparametric regression approach with a multivariable Fourier series estimator. This approach is used when the form of the relationship pattern is unknown and tends to have recurring patterns. The Fourier series estimator depends on the number of oscillations, so in this study, 1 to 4 oscillations were tried, where the minimum GCV value determined the optimal oscillation. The best model was obtained on the analysis results, producing the smallest GCV value, namely the model with 3 oscillations with a GCV value of 1.027. The results of simultaneous and partial hypothesis testing showed that all predictor variables used in this research were proven to influence the Average Length of Schooling. This is also supported by the coefficient of determination value of 85.464%.
Copyrights © 2024