Claim Missing Document
Check
Articles

Found 32 Documents
Search

PENDEKATAN MAZIMUM PENALIZED LIKELIHOOD UNTUK MENGESTIMASI FUNGSI BASELINE HAZARD PADA MODEL COX: STUDI KASUS PASIEN KANKER PAYUDARA Edina, Almira Ivah; Purnami, Santi Wulan; Sukur, Edi; Saputri, Prilyandari Dina; Febrisutisyanto, Ady; Alfajriyah, Aimmatul Ummah
EPSILON: JURNAL MATEMATIKA MURNI DAN TERAPAN Vol 19, No 2 (2025)
Publisher : Mathematics Study Program, Faculty of Mathematics and Natural Sciences, Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20527/epsilon.v19i2.17087

Abstract

Survival analysis is a statistical method that focuses on time-to-event variables, where the event time represents the duration a patient survives during the observation period. This study applies survival analysis to examine factors influencing the survival of breast cancer patients who are receiving treatment at C-Tech Labs Edwar Technology. The data used are right-censored survival data, referring to patients who either survived until the end of the observation period or died from unrelated causes. Risk factors analyzed include age, gender, and cancer stage, while treatment factors consist of surgery, chemotherapy, radiotherapy, and Frequency of Electro Capacitive Cancer Therapy (ECCT) usage. The Cox Proportional Hazard (PH) model combined with the Maximum Penalized Likelihood (MPL) method is used to analyze the effect of these factors on mortality risk, as well as to estimate regression coefficients and the baseline hazard function more accurately. The results indicate that age, frequency of ECCT use, and the status of additional therapies significantly affect patient survival. Older age increases the risk of death, while a higher frequency of ECCT use and the use of additional therapies help reduce that risk. Routine use of ECCT is shown to contribute to extending the survival time of breast cancer patients at C-Tech Labs Edwar Technology, Tangerang. However, potential confounding variables not examined in this study should be considered when interpreting the findings.
QUANTILE BASED PLS-SEM WITH WILD BOOTSTRAP Balami, Abdul Malik; Otok, Bambang Widjanarko; Purnami, Santi Wulan
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 20 No 2 (2026): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol20iss2pp1775-1790

Abstract

Partial Least Squares SEM (PLS-SEM) is the recommended technique for structural equation modeling (SEM), which assesses correlations between latent components concurrently, particularly for small samples and non-normal data. But because traditional PLS-SEM only calculates average correlations between constructs, it runs the risk of overlooking variances in the quantile distribution. Consequently, the creation of the Quantile PLS-SEM approach, which incorporates quantile regression, provides a means to examine correlations across the entire data distribution. To improve estimation, wild bootstrap is used to address heteroscedasticity issues and produce more reliable inferences. The purpose of this study is to develop and apply Quantile based PLS-SEM with Wild Bootstrap to analyze the gizi data status of the Indonesian population based on the Survey Status Gizi Indonesia 2024. The analysis's findings indicate that specific and sensitive interventions have a significant impact on the gizi status of different quantities.