Pemodelan Regresi Nonparametrik Spline Truncated pada Data Angka Kematian Ibu di Jawa Tengah

Qolbi, Latifatul (2026) Pemodelan Regresi Nonparametrik Spline Truncated pada Data Angka Kematian Ibu di Jawa Tengah. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201112_latifatul qolbi-laporan TA br finish (1).docx - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Angka Kematian Ibu (AKI) merupakan salah satu indikator utama derajat kesehatan masyarakat yang mencerminkan kualitas pelayanan kesehatan ibu selama masa kehamilan, persalinan, hingga nifas. Penelitian ini bertujuan untuk memodelkan AKI pada kabupaten/kota di Provinsi Jawa Tengah tahun 2023 menggunakan regresi nonparametrik Spline Truncated linier dengan mengecualikan Kota Magelang karena keterbatasan data. Variabel prediktor yang digunakan adalah persentase peserta KB aktif (X₁), persentase komplikasi kebidanan (X₂), persentase cakupan kunjungan ibu hamil K4 (X₃), persentase persalinan yang ditolong tenaga kesehatan (X₄), dan persentase cakupan kunjungan ibu hamil K1 (X₅). Pemilihan titik knot optimal dilakukan menggunakan metode Generalized Cross Validation (GCV). Hasil penelitian menunjukkan bahwa model terbaik diperoleh pada kombinasi titik knot (1,3,3,1,3), yaitu satu titik knot pada X₁, tiga titik knot pada X₂, tiga titik knot pada X₃, satu titik knot pada X₄, dan tiga titik knot pada X₅, dengan nilai GCV minimum sebesar 858,621. Model terbaik menghasilkan koefisien determinasi sebesar 83,1%, yang menunjukkan bahwa 83,1% variasi AKI antar kabupaten/kota di Provinsi Jawa Tengah dapat dijelaskan oleh variabel prediktor beserta komponen Spline Truncated-nya. Berdasarkan pengujian parameter secara parsial, variabel yang berpengaruh signifikan terhadap AKI adalah X₁, X₂, X₃, dan X₅, sedangkan X₄ tidak berpengaruh signifikan. Dengan demikian, regresi nonparametrik Spline Truncated linier mampu memodelkan pola hubungan AKI secara fleksibel.
===================================================================================================================================
The Maternal Mortality Ratio (MMR) is one of the primary indicators of public health status, reflecting the quality of maternal healthcare services during pregnancy, childbirth, and the postpartum period. This study aims to model the MMR across districts and municipalities in Central Java Province in 2023 using linear nonparametric Truncated Spline regression, excluding Magelang City due to data limitations. The predictor variables include the percentage of active family planning participants (X₁), the percentage of obstetric complications (X₂), the percentage of fourth antenatal care (K4) coverage (X₃), the percentage of deliveries attended by skilled health personnel (X₄), and the percentage of first antenatal care (K1) coverage (X₅). The optimal knot points were determined using the Generalized Cross Validation (GCV) method. The results indicate that the best model was obtained using the knot combination (1,3,3,1,3), consisting of one knot for X₁, three knots for X₂, three knots for X₃, one knot for X₄, and three knots for X₅, with a minimum GCV value of 858.621. The best model achieved a coefficient of determination of 83.1%, indicating that 83.1% of the variation in MMR among districts and municipalities in Central Java Province can be explained by the predictor variables and their Truncated Spline components. Based on the partial parameter significance tests, the variables that significantly influence MMR are X₁, X₂, X₃, and X₅, whereas X₄ does not have a significant effect. Therefore, linear nonparametric Truncated Spline regression is capable of modeling the relationship pattern of MMR in a flexible manner.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Angka Kematian Ibu, Jawa Tengah, Regresi Nonparametrik,Spline Truncated Maternal Mortality Ratio, Central Java, Nonparametric Regression, Truncated Spline.
Subjects:	Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Latifatul Qolbi
Date Deposited:	06 Aug 2026 02:39
Last Modified:	06 Aug 2026 02:39
URI:	http://repository.its.ac.id/id/eprint/144131

Actions (login required)

View Item