NATALIE EFRATA SUSANTI, . (2025) PENANGANAN MULTIKOLINEARITAS MENGGUNAKAN REGRESI KUADRAT TERKECIL PARSIAL DAN REGRESI KOMPONEN UTAMA PADA PREVALENSI STUNTING DI PROVINSI NUSA TENGGARA TIMUR 2022. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.
![]() |
Text
COVER.pdf Download (2MB) |
![]() |
Text
BAB I.pdf Download (574kB) |
![]() |
Text
BAB II.pdf Restricted to Registered users only Download (768kB) | Request a copy |
![]() |
Text
BAB III.pdf Restricted to Registered users only Download (876kB) | Request a copy |
![]() |
Text
BAB IV.pdf Restricted to Registered users only Download (731kB) | Request a copy |
![]() |
Text
BAB V.pdf Restricted to Registered users only Download (736kB) | Request a copy |
![]() |
Text
DAFTAR PUSTAKA.pdf Download (574kB) |
![]() |
Text
LAMPIRAN.pdf Restricted to Registered users only Download (896kB) | Request a copy |
Abstract
Permasalahan gizi menjadi penyebab sekitar setengah dari kematian anak usia balita, terutama di negara berekonomi rendah dan sedang. Di Indonesia, stunting adalah salah satu masalah bidang gizi yang banyak terjadi, yakni ketika tinggi badan anak jauh di bawah rata-rata anak seusianya. Pada tahun 2022, Nusa Tenggara Timur menempati peringkat teratas prevalensi stunting di Indonesia, yakni sekitar 35,3%. Tujuan dari penelitian ini adalah membandingkan metode regresi kuadrat terkecil parsial (RKTP) algoritma NIPALS dengan regresi komponen utama (RKU) pada penanganan permasalahan multikolinearitas. Penelitian ini memakai data sekunder dari publikasi hasil Survei Status Gizi Indonesia (SSGI) tahun 2022 oleh Kementerian Kesehatan dan Badan Pusat Statistik NTT. Data yang digunakan terdiri atas satu variabel respon dan 10 variabel prediktor. Hasil analisis penelitian menyatakan bahwa model RKTP lebih unggul daripada RKU. Hal ini ditunjukkan oleh nilai Adj.R2 RKTP sebesar 0,741 yang lebih besar daripada RKU yang hanya sebesar 0,322. Selanjutnya, lebih jauh kebaikan RKTP berdasarkan nilai RMSE dan MAE berturut-turut 2,783 dan 1,910. Nilai tersebut lebih rendah jika dibandingkan pada nilai RMSE dan MAE RKU berturut-turut 4,742 dan 3,346. Berdasarkan model dugaan RKTP, ada lima variabel prediktor yang memiliki pengaruh signifikan pada variabel respon, yaitu rata-rata konsumsi protein per kapita sehari (X2), jumlah balita yang pernah mendapatkan imunisasi DPT dan HB (X3), Indeks Pembangunan Manusia (X7), persentase rumah tangga berdasarkan sumber air minum layak (X8), serta jumlah penduduk miskin (X9). ***** Nutrition problems are the cause of about half of all deaths among children under five, especially in low- and middle-income countries. In Indonesia, stunting is one of the most prevalent nutrition problems, which occurs when a child’s height is far below the average for children of the same age. In 2022, East Nusa Tenggara ranked highest in Indonesia for stunting prevalence, at around 35,3%. The objective of this study is to compare the partial least squares regression (PLSR) algorithm using the NIPALS method with principal component regression (PCR) in addressing multicollinearity issues. This study utilizes secondary data from the 2022 Indonesia Nutrition Status Survey (SSGI) published by the Ministry of Health and the Central Statistics Agency of East Nusa Tenggara. The data used consists of one response variable and 10 predictor variables. The results of the analysis indicate that the PLSR model is superior to PCA. This is demonstrated by the adjusted R2 value of PLSR at 0,741, which is significantly higher than PCR’s value of 0,322. Furthermore, the superiority of PLSR is further supported by the RMSE and MAE values of 2,783 and 1,910, respectively. These values are lower compared to the RMSE and MAEvalues of PCR, which are 4,742 and 3,346, respectively. Based on the PLSR prediction model, there are five predictor variables that have a significant influence on the response variable, namely the average daily protein consumption per capita (X2), the number of infants who have received DPT and HB immunizations (X3), the Human Development Index (X7), the percentage of households based on access to safe drinking water (X8), and the number of poor people (X9).
Item Type: | Thesis (Sarjana) |
---|---|
Additional Information: | 1.) Dr. Vera Maya Santi, S.Si., M.Si. 2.) Devi Eka Wardani Meganingtyas, S.Pd., M.Si. |
Subjects: | Sains > Matematika |
Divisions: | FMIPA > S1 Matematika |
Depositing User: | Natalie Efrata Susanti . |
Date Deposited: | 12 Aug 2025 04:07 |
Last Modified: | 12 Aug 2025 04:07 |
URI: | http://repository.unj.ac.id/id/eprint/60098 |
Actions (login required)
![]() |
View Item |