Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Journal of Mathematics, Computation and Statistics (JMATHCOS)
Isti Kamila
Institut Teknologi Kalimantan
Mathematics

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Classification of Money Market Mutual Fund Products in Indonesia By Using Mahalanobis Distance and Manhattan Distance Indrawan; Muhammad Azka; Isti Kamila; Nurul Maqfirah Rauf; Eka Krisna Santoso
Journal of Mathematics, Computations and Statistics Vol. 8 No. 1 (2025): Volume 08 Nomor 01 (April 2025)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v8i1.6143

Abstract

This study aims to classify money market fund products listed and supervised by the Financial Services Authority (OJK) with minimal classification error. Mahalanobis distance and Manhattan distance were employed to classify these products. Data was sourced from the Indo Premier Online Technology (IPOT) application. Variables utilized in this research include percentage return, Sharpe ratio, unit growth, and Asset Under Management (AUM) . Additionally, Principal Component Analysis (PCA) was employed to reduce data dimensionality by linearly combining correlated original variables into new variables (principal components). PCA was used to visualize data with more than three dimensions. Based on the principal component analysis, the first two principal components captured 74.43% of the original data information, while the first three principal components captured 98.94%. Classification results using three principal components and standardized data showed the same error rates: 13.33% for Mahalanobis distance and 6.67% for Manhattan distance. For the two principal components, both Mahalanobis and Manhattan distances resulted in an error rate of 13.33%. Therefore, Manhattan distance is the most effective method for classification. Forecasting results indicate that mutual fund A is a good investment choice, while mutual fund B is a poor one. Keywords: Mahalanobis distance; Manhattan distance; Principal Component Analysis
Mortality Risk Modeling for Patients with Lung Disease Using Cox Regression and Gradient Boosting Survival at Pertamina Hospital in Balikpapan Nurul Maqfirah Rauf; Isti Kamila; Muhammad Azka; Indrawan Indrawan; Nur Qadri Bahar
Journal of Mathematics, Computations and Statistics Vol. 9 No. 3 (2026): Volume 09 Issue 03 (September 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/137g3361

Abstract

Lung disease is a health issue associated with mortality risk; therefore, an analysis is needed to identify factors associated with the risk of death and to model patient survival risk. This study aims to identify factors associated with mortality risk in patients with lung disease using Cox Proportional Hazards Regression and to develop a predictive model using Gradient Boosting Survival. The study utilized medical records from 1,119 inpatients with lung disease at Pertamina Hospital in Balikpapan during the 2024–2025 period. Analyses were conducted using Kaplan-Meier, Cox Proportional Hazards Regression, and Gradient Boosting Survival. Hyperparameter optimization for Gradient Boosting Survival was performed using Grid Search with 5-fold stratified cross-validation on the training data. The results of the Cox Proportional Hazards Regression showed that BMI, diastolic blood pressure, and Chronic Obstructive Pulmonary Disease (COPD) were significantly associated with the risk of mortality, with a C-index of 0.6500. Gradient Boosting Survival yielded the optimal hyperparameters: a learning rate of 0.10, n_estimators of 200, max_depth of 2, and min_samples_leaf of 10, with a C-index of 0.6576 on the training data and 0.5575 on the testing data. Feature importance analysis shows that BMI, diastolic blood pressure, systolic blood pressure, SpO₂, and COPD make the largest contributions to the model. Cox Proportional Hazards Regression provides information on the direction and magnitude of the relationship between risk factors through the hazard ratio, while Gradient Boosting Survival provides information on the relative contributions of variables through feature importance. Thus, both approaches provide complementary information in modeling the risk of mortality in patients with lung disease.
Co-Authors Eka Krisna Santoso Indrawan Muhammad Azka Muhammad Azka Nur Qadri Bahar Nurul Maqfirah Rauf