Claim Missing Document
Check
Articles

Found 10 Documents
Search

Analisis Ridge Robust Penduga Generalized M (GM) Pada Pemodelan Kalibrasi Untuk Kadar Gula Darah Agung Tri Utomo; Erfiani, Erfiani; Fitrianto, Anwar
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 2 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (384.666 KB) | DOI: 10.35580/variansiunm14

Abstract

Calibration modeling is one of the methods used to analyze the relationship between different methods. The relationship is like the relationship between invasive and non-invasive blood sugar measurement. Problems that often arise in calibration modeling are multicollinearity and outliers. Multicollinearity problems can cause the regression confidence interval to widen, so that there is no statistically significant regression coefficient. Outliers cause statistical tests to deviate. The handling of these problems can be solved by robust ridge analysis. Ridge robust is a combined analysis of ridge regression and robust regression. Ridge regression is able to overcome the problem of multicollinearity and robust regression can overcome the problem of outliers. The estimator used is Generalized M (GM). This method will be applied to a calibration model that uses invasive and non-invasive blood sugar level data. The model used with Generalized M (GM) estimator robust regression using modulation clusters 50 to 90 in 2017 is better than the modulation group 50. up to 90 in 2019. The statistical values obtained are SSE of 0.910, RMSEadj of 0.114, and RMSEP of 0.030. Calibration models that have outliers and multicollinearity problems can be overcome by robust ridge regression. The feasibility value of the model obtained in the GM estimator robust regression is smaller than the MM estimator ridge robust regression in the calibration modeling for non-invasive blood sugar level data. That is, the best model that can be used is the robust ridge regression GM estimator.
TSA App by R Shiny : Time Series Analysis Application for Univariate Series Data Tri Utomo, Agung; Ahmar, Ansari Saleh; Aidid, Muhammad Kasim; Rais, Zulkifli; Alfairus, Muh. Qodri
ARRUS Journal of Engineering and Technology Vol. 5 No. 1 (2025)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/jetech4398

Abstract

Time series analysis is a statistical method used to model and forecast sequential data over time. This modeling is typically performed using software, but most analytical tools require paid licenses. To address this issue, the TSA App by R Shiny is developed as an open-source application that is easily accessible. The application features a dashboard-based interface designed to help users perform univariate time series analysis without requiring programming skills. This study compares the analysis results of the TSA App with other software such as R Studio, Minitab, and Python. The results show that the TSA App produces comparable outputs in terms of visualization, ARIMA modeling, and forecasting accuracy. Therefore, the TSA App provides a practical and legal solution for time series analysis, especially for users who are unfamiliar with coding.
Optimizing Fraud Detection in Indonesia via Rare-Event Logit Approach: A Simulation Study on Large-Scale Agung Tri Utomo; Abdul Rahman
Journal of Economics, Entrepreneurship, Management Business and Accounting Vol 4 No 1 (2026): Volume 4, Issue 1, January 2026
Publisher : CV. Sakura Digital Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61255/jeemba.v4i1.1309

Abstract

Purpose: This study examines the use of the Rare-Event Logit approach to improve fraud detection under conditions of extreme class imbalance. The topic is important because fraud cases usually represent only a very small proportion of total financial transactions, which may reduce the accuracy of conventional classification models. Design/methodology/approach: This study uses a simulation-based quantitative design to evaluate fraud detection performance in large-scale imbalanced data settings. The analysis compares standard logistic regression and Rare-Event Logit with bias-corrected estimation, including Firth’s penalized likelihood approach. Model performance is assessed using the Area Under the Precision-Recall Curve and F1-Score. Findings/Results: The results show that standard logit and Rare-Event Logit perform similarly under moderate imbalance conditions. However, Rare-Event Logit provides a stronger theoretical advantage in handling rare-event bias and stabilizing parameter estimation as data sparsity increases. This indicates that bias-corrected probabilistic models are more suitable for fraud detection in highly imbalanced environments. Originality/Value: This study highlights the value of Rare-Event Logit as an alternative approach for fraud detection in rare-event settings. The findings imply that financial institutions can improve fraud risk identification by adopting bias-corrected models that are more robust to class imbalance.
Forecasting IDR Exchange Rate to USD Using Hybrid ARIMA – LSTM Zulkifli Rais; Sitti Masyitah Meliyani R; Astrid Suwardani Sumarno; Agung Tri Utomo
ARRUS Journal of Engineering and Technology Vol. 6 No. 1 (2026)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/jetech4899

Abstract

Time series forecasting often involves both linear and nonlinear patterns, making the use of a single method less effective. This study aims to forecast the exchange rate of the Indonesian Rupiah (IDR) against the United States Dollar (USD) using a hybrid ARIMA–LSTM model. ARIMA is used to capture linear patterns, while LSTM is employed to model nonlinear residual components. The data used are weekly exchange rates from January 2020 to August 2025. Model performance is evaluated using Mean Absolute Percentage Error (MAPE). The results show that the hybrid ARIMA–LSTM model produces better forecasting accuracy compared to individual ARIMA and LSTM models, with the lowest MAPE value of 0.73%. This indicates that combining linear and nonlinear modeling approaches improves forecasting performance for complex time series data.
Application of LASSO Regression for the Identification of Underdeveloped Regions in Central Sulawesi Muh. Qodri Alfairus; Husnul Amira; Agung Tri Utomo; Nur Abshari Abbas
ARRUS Journal of Mathematics and Applied Science Vol. 6 No. 1 (2026)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/mathscience4813

Abstract

This study aims to identify the main factors influencing regional underdevelopment in Central Sulawesi through Human Development Index (HDI) modeling and to develop a robust predictive model. To address the challenges of multicollinearity and the limited number of observations (13 districts/cities with 10 variables), this study employs LASSO (Least Absolute Shrinkage and Selection Operator) regression, which is capable of simultaneously shrinking coefficients and selecting variables. The data used are sourced from the 2019 publication of the Central Statistics Agency (BPS). The analysis was conducted using descriptive statistics, Ordinary Least Squares (OLS) modeling, VIF tests, and LASSO regression with cross-validation (leave-one-out cross-validation). The results indicate that very high multicollinearity (VIF > 10 for most variables) renders the OLS model unstable. Conversely, LASSO regression yielded better performance with superior RMSE (1.282), MAE (1.075), and R² (0.918) values compared to OLS (RMSE 21.67; MAE 9.85; R² 0.78). Thus, LASSO is more suitable for limited data with high multicollinearity. The selected significant variables include the percentage of the poor population, the open unemployment rate, shopping facilities, the presence of hospitals, the population density ratio, and the number of elementary and secondary schools.
Factors Associated with Monocyte and SuPAR Levels in Pulmonary Tuberculosis Patients: A Longitudinal Study Lalu Ramzy Rahmanda; Agung Tri Utomo; Nur Ikhwana; Annisa Syalsabila
ARRUS Journal of Mathematics and Applied Science Vol. 6 No. 1 (2026)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/mathscience4851

Abstract

Tuberculosis in Indonesia had the second highest TB burden globally, after India. 92% of the estimated cases were confirmed to be pulmonary tuberculosis. However, controlling pulmonary tuberculosis relies heavily on accurate diagnosis, appropriate treatment, effective monitoring, and evaluation. This study aims to model and investigate the most significant factor affecting monocyte and suPAR levels as a biomarker based on observation time, body mass index, and erythrocyte sedimentation rate. This study was a longitudinal study. 60 patients were involved every two weeks over 13 periods. Path analysis with the generalized linear mixed model (GLMM) approach and comparing three estimation methods to investigate the longitudinal relationship between variables and compare the best structure for modeling the relationship. The best model for describing the relationship between observation time, body mass index, and erythrocyte sedimentation rate on monocyte and suPAR levels is GLMM with unstructured covariance (R2 = 0.977 and AIC = 0.052). A significant positive correlation between monocyte and suPAR levels further validates suPAR as a robust biomarker for monitoring treatment response. The study concludes that effective clinical management of pulmonary tuberculosis requires an integrated strategy that combines OAT with regular monitoring of BMI, ESR, and monocyte levels to optimize patient recovery.
Workshop on Student Graduation Decisions Using Statistical Methods at Takalar State Senior High School 7 Suwardi Annas; Ansari Saleh Ahmar; Zulkifli Rais; Rahmat H.S; Agung Tri Utomo
ARRUS Jurnal Pengabdian Kepada Masyarakat Vol. 4 No. 2 (2025)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/454RI.abdiku4458

Abstract

This community service program was conducted at SMA Negeri 7 Takalar to enhance teachers’ ability to utilize statistical methods specifically logistic regression to support data-driven graduation decisions. The training addressed challenges related to manual graduation assessment processes that often lack objective analytical support. Participants were introduced to the basic concepts of logistic regression, followed by hands-on practice using an interactive R Shiny dashboard to analyze student data and estimate graduation probabilities. The results indicate that teachers were able to understand and apply statistical analysis procedures, interpret logistic regression outputs, and recognize the importance of evidence-based decision-making. This activity not only improved teachers’ data literacy but also supported digital transformation efforts in education and strengthened collaboration between Universitas Negeri Makassar and SMA Negeri 7 Takalar. The program is expected to contribute to more accurate, transparent, and data-informed graduation assessments in the future.
Temporal Aggregation and Smoothing Parameter Sensitivity in Exponential Smoothing: Evidence from the Jakarta Composite Index Agung Tri Utomo; Abdul Rahman; Zulkifli Rais; Muh. Qodri Alfairus
Daengku: Journal of Humanities and Social Sciences Innovation Vol. 5 No. 6 (2025)
Publisher : PT Mattawang Mediatama Solution

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/454RI.daengku4925

Abstract

The study examines the structural sensitivity and predictive performance of the Single Exponential Smoothing (SES) model when applied to varying temporal aggregation levels of the Indonesian stock market index. This study employs time-series data of the Jakarta Composite Index (JCI / IHSG) spanning from January 2, 2020, to May 31, 2025, which are structurally categorized into three distinct frequency domains: daily observations, weekly aggregated averages, and monthly aggregated averages. Methodologically, the optimal smoothing parameter (?) for each aggregation tier is determined through maximum likelihood estimation by minimizing the Mean Absolute Percentage Error (MAPE). The results reveal an extreme structural behavior where the optimal ? approaches its upper asymptotic boundary across all temporal frameworks, specifically 0.9943 for daily data, 0.9999 for weekly data, and 0.9999 for monthly data. Concurrently, the predictive error amplifies systematically as the aggregation window widens, yielding MAPE values of 0.74%, 1.30%, and 2.88% for daily, weekly, and monthly frameworks, respectively. The findings of this study suggest that the JCI movement exhibits strong adherence to the random walk hypothesis, wherein the mathematical framework of SES reacts almost exclusively to the most recent historical innovation. Consequently, temporal aggregation does not induce a structural smoothing effect on parameter convergence but rather introduces informational attenuation that compromises short-term forecasting precision.
Earth System Variables as Drivers of Environmental Commodity Price Dynamics: A Systematic Review of Physics-Informed and Data-Driven Modelling Approaches (2010–2026) Muhammad Nusrang; Ansari Saleh Ahmar; Abdul Rahman; Agung Tri Utomo; Muh. Qodri Alfairus
ARRUS Journal of Social Sciences and Humanities Vol. 6 No. 3 (2026)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/soshum4973

Abstract

Environmental commodity markets (carbon allowances, electricity, natural gas, and renewable energy instruments) are, at their root, earth system markets: their price-generating processes are as much a product of atmospheric circulation, hydrological regimes, and climate teleconnections as of supply-demand fundamentals or regulatory signals. Despite this physical reality, the machine learning forecasting literature has treated these markets primarily as benchmarking arenas, decoupling predictive architectures from the physical processes that drive price-relevant forcing. This systematic review follows PRISMA 2020 guidelines on a corpus of 413 peer-reviewed studies drawn from 652 Scopus records (2010–2026) and examines how earth system variables, including temperature anomalies, precipitation regimes, wind resource indices, atmospheric pollution metrics, and climate teleconnections such as ENSO and NAO, have been integrated into ML-based environmental commodity price models, and evaluates the evidence for whether physics-informed feature engineering confers measurable accuracy advantages over purely data-driven approaches. Bibliometric analysis reveals rapid field expansion (71.6% of publications in 2021–2026), geographic concentration in Chinese ETS research (?68% of high-impact output), and methodological dominance of hybrid decomposition-deep learning architectures. Models incorporating earth system variables consistently outperform endogenous-only ML architectures by 12–35% on MAPE, yet fewer than 30% of corpus studies include any physical predictor and climate teleconnection indices appear in under 4% of studies despite their relevance at energy market planning horizons. Six research priorities are identified, centred on numerical weather prediction ensemble integration, cross-climate-regime validation, and probabilistic forecasting grounded in physical uncertainty quantification.
Land Cover Change in the Mamminasata Metropolitan Area, Indonesia, 2005–2025 and CA-Markov Prediction for 2035 Sitti Masyitah Meliyana; Agung Tri Utomo; Muh. Qodri Alfairus
ARRUS Journal of Mathematics and Applied Science Vol. 6 No. 2 (2026)
Publisher : PT ARRUS Intelektual Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35877/mathscience5506

Abstract

This study examines land cover change in the Mamminasata Metropolitan Area (Makassar, Maros, Gowa, and Takalar), South Sulawesi, Indonesia, between 2005 and 2025 and projects its land cover to 2035. Landsat 5 TM and Landsat 8/9 OLI surface reflectance composites were classified into seven classes on Google Earth Engine using a Random Forest classifier trained with samples derived automatically from Dynamic World labels, followed by rule-based post-classification and temporal consistency corrections. Future land cover was simulated with a CA-Markov model that combines Markov transition quantities, a gradient boosting suitability model, and cellular automata allocation. The classified maps reached overall accuracies of 77.9% (2005), 81.6% (2015), and 84.9% (2025), and the 2025 hindcast agreed with the observed map at 87.3% (Kappa 0.82; Figure of Merit 0.13). Built-up land expanded from 38,036 ha to 60,661 ha (+59.5%), while paddy fields declined by 9,457 ha (?15.1%). Dryland farms and shrubland supplied 55.3% and paddy fields 29.0% of the new built-up land in 2015–2025, and 61% of the 2005–2025 growth occurred more than 20 km from the city centre. Under a business-as-usual trend, built-up land will reach 71,293 ha in 2035, mainly within 5–15 km of the city centre, and paddy fields will fall to 48,924 ha. The findings indicate that urban growth in Mamminasata has shifted from the core city to its agricultural periphery.