http://eprints.unpak.ac.id/id/eprint/11067 2026-07-29T01:50:31Z info:eu-repo/semantics/descriptiveMetadata http://eprints.unpak.ac.id/11067/ ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI KEJADIAN BALITA STUNTING DI INDONESIA MENGGUNAKAN REGRESI SEMIPARAMETRIK SPLINE Ocsiella, Saskia Yety Matematika ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI KEJADIAN BALITA STUNTING DI INDONESIA MENGGUNAKAN REGRESI SEMIPARAMETRIK SPLINE Sakia Yety Ocsiella*, Yasmin Erika Faridhan, Hagni Wijayanti FMIPA, Universitas Pakuan, Jl. Pakuan, RT.02/RW.06, Tegallega, Kecamatan Bogor Tengah, Kota Bogor, Jawa Barat 16129 e-mail: sakiayocsiella@gmail.com Abstract: This study aims to model the percentage of stunting and to analyze the factors influencing stunting prevalence in Indonesia. The explanatory variables considered include the percentage of animal-source complementary feeding (MPASI), child nutritional status (weight-for-age), access to improved sanitation, access to unimproved drinking water, and the percentage of pregnant women consuming iron supplementation (TTD). The scatter plots between the response and explanatory variables indicate nonlinear patterns; therefore, a semiparametric spline regression approach is employed. Semiparametric spline regression offers flexibility in capturing both linear and non-linear relationships through the use of knot points. In this study, models are constructed using one, two, three, four, and a combination of knot points. The optimal model is selected based on the minimum value of Generalized Cross Validation (GCV). The data used are secondary data obtained from the Ministry of Villages, Development of Disadvantaged Regions, and Transmigration, covering 38 provinces in Indonesia in 2023. The results show that the best model is achieved using a combination of knot points (2,3,4), with a minimum GCV value of 30.5424 and a coefficient of determination (R²) of 89.42%. The findings indicate that all explanatory variables significantly influence the percentage of stunting. In conclusion, the semiparametric spline regression model is effective in modeling stunting prevalence in Indonesia and provides a flexible approach to capturing complex relationships between variables. Keywords: stunting, semiparametric regression, spline, GCV, explanatory variables 2025 Thesis NonPeerReviewed Ocsiella, Saskia Yety (2025) ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI KEJADIAN BALITA STUNTING DI INDONESIA MENGGUNAKAN REGRESI SEMIPARAMETRIK SPLINE. Skripsi thesis, Universitas Pakuan. info:eu-repo/semantics/humanStartPage