Claim Missing Document
Check
Articles

Evaluasi Performa Model Regresi Poisson Tweedie dan Conway Maxwell Poisson dalam Menangani Masalah Dispersi: Studi Angka Kematian Ibu di Provinsi Sulawesi Selatan Aswi, Aswi; Sanusi, Wahidah; Tiro, Muhammad Arif; Sukarna, Sukarna; Haekal, Muh. Fahri; Palarungi, Andi Gagah; Putri, Siti Choirotun Aisyah; Oktaviana, Oktaviana
Indonesian Journal of Fundamental Sciences Vol 11, No 2 (2025)
Publisher : Universitas Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26858/ijfs.v11i2.77506

Abstract

Model regresi Poisson digunakan untuk menganalisis hubungan antara satu atau lebih variabel independen dengan variabel dependen berupa data cacahan. Salah satu asumsi utamanya adalah kesamaan antara nilai mean dan variansi (equidispersi). Namun, dalam praktiknya, asumsi tersebut sering tidak terpenuhi. Kondisi ini menyebabkan model regresi Poisson kurang sesuai digunakan, karena dapat menghasilkan estimasi standar error yang terlalu kecil (underestimate). Model alternatif yang dapat digunakan untuk mengatasi masalah overdispersi adalah Regresi Poisson Tweedie dan Conway Maxwell Poisson (CMP). Penelitian ini bertujuan untuk mengevaluasi kinerja model regresi Poisson Tweedie dan regresi CMP dalam menangani masalah dispersi pada data Angka Kematian Ibu (AKI) di Provinsi Sulawesi Selatan, Indonesia. Estimasi parameter dilakukan dengan metode Estimasi Kemungkinan Maksimum (MLE), sedangkan kinerja model dinilai berdasarkan Akaike Information Criterion (AIC), Mean Square Error (MSE), dan signifikansi parameter. Hasil penelitian menunjukkan bahwa model regresi Poisson standar kurang sesuai karena adanya pelanggaran asumsi ekuidispersi. Sebaliknya, model CMP dan Poisson Tweedie memberikan alternatif yang lebih tepat, dimana Model CMP menunjukkan akurasi prediktif yang lebih tinggi dengan nilai MSE terendah. Faktor perdarahan, hipertensi, gangguan kardiovaskular, dan komplikasi pasca-aborsi ditemukan memiliki pengaruh yang signifikan terhadap kematian ibu, sementara infeksi tidak signifikan secara statistik. 
The Application of the K-Medoid Classification Method for Analyzing Crime Rates in South Sulawesi Annas, Suwardi; Aswi, Aswi; Irwan, Irwan
Inferensi Vol 8, No 3 (2025)
Publisher : Department of Statistics ITS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j27213862.v8i3.21464

Abstract

This research employs the k-medoid clustering method to analyze districts and cities in South Sulawesi based on their crime rates. As the population grows, employment opportunities tend to diminish, which can increase stress levels and, consequently, the likelihood of criminal behavior. To evaluate the distribution of criminal incidents across South Sulawesi, the k-medoid method is used to cluster regions. Unlike other clustering methods, k-medoid utilizes the median as the cluster center (medoid), which enhances its robustness against outliers. Specifically, the Partitioning Around Medoids (PAM) algorithm is applied, where initial objects are randomly selected to represent clusters. If the error value is high, the cluster centers are adjusted until the error is minimized. The dataset comprises crime incidence data for South Sulawesi in 2020, focusing on various types of crime. The analysis identified an optimal number of three clusters based on the Silhouette coefficient. Cluster 1 includes 11 regions, Cluster 2 consists of 8 regions, and Cluster 3 contains 5 regions. These clusters provide a comprehensive overview of the crime conditions across different regions within each cluster.
Bayesian Spatio-Temporal Conditional Autoregressive Modelling of Factors Affecting Pneumonia Cases in Indonesia Risma Mastory; Aswi, Aswi; Muhammad Fahmuddin; Lalu Ramzy Rahmanda
Jurnal MSA (Matematika dan Statistika serta Aplikasinya) Vol 13 No 2 (2025): VOLUME 13 NO 2, 2025
Publisher : Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/msa.v13i2.56315

Abstract

The Bayesian Spatio-Temporal Conditional Autoregressive (BST CAR) method is a statistical approach used to analyze data with both spatial and temporal components. While the BST CAR model has been widely applied in various studies, no research has yet explored using the Localized BST CAR model for pneumonia cases in Indonesia. This study aims to identify and model the factors influencing pneumonia incidence in Indonesia using the Localized BST CAR framework. The data analyzed in this study consist of the number of pneumonia cases in Indonesia from 2018 to 2022, along with variables believed to affect the incidence. The findings indicate that the Localized BST CAR model with G=3 provides the best fit for modeling the relative risk of pneumonia cases in Indonesia. Key factors found to significantly influence pneumonia cases include the percentage of exclusively breastfed infants, the percentage of infants with complete basic immunization, and the percentage of the population living in poverty. Notably, the percentage of exclusively breastfed infants and the percentage of fully immunized infants were positively associated with pneumonia cases, while the percentage of the poor population had a negative effect
Estimating the Relative Risk of Dengue Hemorrhagic Fever in Makassar City Using a Bayesian Spatial Localised Conditional Autoregressive Model Rahmawati; Aswi Aswi; Rahmat Hidayat; Andi Gagah Palarungi Taufik
Journal of Mathematics, Computations and Statistics Vol. 9 No. 1 (2026): Volume 09 Issue 01 (March 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/z0mxxw06

Abstract

Dengue Hemorrhagic Fever (DHF) remains a significant public health challenge in Indonesia, including in Makassar City, which reported an increase of 291 cases in 2024. This study aimed to estimate the relative risk of DHF across 15 districts of Makassar by incorporating covariates such as population density, distance to the city center, and the number of hospitals, using a Bayesian Conditional Autoregressive (CAR) Localised approach. The data were obtained from the publication Makassar City in Figures 2025, issued by the Central Statistics Agency. Spatial autocorrelation analysis with Moran’s I indicated significant clustering of DHF cases. Model selection was conducted using the Deviance Information Criterion (DIC), Watanabe–Akaike Information Criterion (WAIC), and group-level area coverage. The results showed that the best-fitting model was the CAR Localised model with distance as a covariate (M9), specified at G = 3 with hyperprior IG (1; 0.01). Distance exhibited a negative association with DHF incidence, suggesting that the farther a district is from the city center, the lower its relative risk. Among the districts, Rappocini exhibited the highest relative risk followed by Panakkukang, while the lowest risks were observed in Sangkarrang Islands. These findings provide valuable insights for designing spatially targeted DHF prevention and control strategies in Makassar City.
Tourism Forecasting Using Chen and Singh Fuzzy Time Series Models Vivianti Vivianti; Aswi Aswi
Journal of Mathematics, Computations and Statistics Vol. 9 No. 1 (2026): Volume 09 Issue 01 (March 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/w2554664

Abstract

The tourism sector is one of the main drivers of the national economy, which experienced a significant decline due to the COVID-19 pandemic. In the post-pandemic era, the recovery of international tourist arrivals shows a positive trend, thus requiring accurate forecasting methods to support tourism policy planning. ARIMA method are less effective in handling nonlinear and fluctuating data. This study applies the Fuzzy Time Series (FTS) approach, specifically the Chen and Singh models, which are capable of managing data uncertainty and representing linguistic patterns adaptively. The purpose of this study is to compare the accuracy of both models using two interval determination approaches, namely the Sturges method and the mean-based method, in forecasting international tourist arrivals through Sultan Hasanuddin International Airport during the period from January 2023 to September 2025. The analytical steps include defining the universe of discourse, performing fuzzification, constructing fuzzy logical relationships (FLR) and fuzzy logical relationship groups (FLRG), and applying defuzzification to obtain forecasted values. The forecasting accuracy was evaluated using the Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The results show that the choice of interval determination method significantly affects forecasting performance, with the mean-based method producing more detailed and accurate intervals. Based on the evaluation, the FTS Singh model demonstrated the best performance, with MAPE of 2.16% and RMSE of 31.05, outperforming the Chen model under both interval approaches. Therefore, the combination of the FTS Singh model with the mean-based interval method is recommended as the optimal approach for forecasting post-pandemic international tourist arrivals, as it can capture fluctuating data patterns more precisely and consistently.
A Systematic Simulation Study of Semiparametric Spline Estimators for Nonlinear Data Structures in R Rahmat Hidayat; Aswi Aswi; Zakiyah Mar'ah
Journal of Mathematics, Computations and Statistics Vol. 9 No. 2 (2026): Volume 09 Issue 02 (June 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/Jmathcos11728

Abstract

In many real-world applications, the assumption of linearity in classical regression models is often violated, leading to model misspecification and inaccurate estimation when data exhibit complex nonlinear patterns. Although nonparametric approaches provide flexibility, they frequently suffer from poor interpretability and instability in high-dimensional settings. To address these limitations, this study examines the implementation of semiparametric spline regression as a flexible yet interpretable alternative. The model integrates a linear component for certain predictors and a spline-based nonparametric component to capture local data fluctuations. Through a simulation study using the R programming language, the performance of the spline estimator was evaluated based on the Generalized Cross Validation (GCV) criterion for optimal knot selection. The results demonstrate that the semiparametric spline model achieves superior accuracy, with a coefficient of determination (R²) reaching 97.35%, compared to 81.18% for the classical linear model. In addition, the Mean Square Error (MSE) is significantly reduced from 2.158 to 0.303. Residual diagnostic analysis confirms that the model satisfies normality and homoscedasticity assumptions. These findings highlight the effectiveness of spline-based semiparametric regression in modeling complex nonlinear data structures.
Comparison of Bayesian Spatial Leroux CAR Models with Poisson and Binomial Likelihoods for Modeling Stunting Cases Rika Saniarti; Aswi Aswi
Journal of Mathematics, Computations and Statistics Vol. 9 No. 2 (2026): Volume 09 Issue 02 (June 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/Jmathcos12090

Abstract

Stunting remains a chronic nutritional problem requiring precise spatial mapping to support effective policy interventions in East Java, Indonesia. Spatial disease mapping commonly applies the Poisson distribution with Conditional Autoregressive (CAR) effects; however, the Poisson distribution is sensitive to overdispersion. Alternatively, the Bayesian Spatial CAR model with a Binomial likelihood may offer a better framework, yet empirical comparisons remain limited. This study compares the performance of Bayesian Spatial Leroux CAR models with Poisson and Binomial likelihoods in modeling stunting cases and identifies associated factors. The data include stunting cases across 38 districts in East Java (2024) and predictors: low birth weight (LBW), prematurity, exclusive breastfeeding, complete basic immunization (CBI), pneumonia, and diarrhea. Performance was evaluated using the Deviance Information Criterion (DIC) and the Watanabe–Akaike Information Criterion (WAIC). Results indicate significant spatial dependence. The Bayesian Spatial Leroux CAR model with a binomial likelihood outperforms the Poisson-based model. LBW, exclusive breastfeeding, CBI, pneumonia, and diarrhea are significantly associated with stunting. Kediri Regency exhibits the highest relative risk (RR), followed by Probolinggo Regency and Batu City, while Kediri City and Ponorogo Regency show the lowest RR
Binary Logistic Regression Model of Stroke Patients: A Case Study of Stroke Centre Hospital in Makassar Suwardi Annas; Aswi Aswi; Muhammad Abdy; Bobby Poerwanto
Indonesian Journal of Statistics and Applications Vol 6 No 1 (2022)
Publisher : Statistics and Data Science Program Study, SSMI, IPB University, in collaboration with the Forum Pendidikan Tinggi Statistika Indonesia (FORSTAT) and the Ikatan Statistisi Indonesia (ISI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/ijsa.v6i1p161-169

Abstract

This paper aimed to determine factors that affect significantly types of stroke for stroke patients in Dadi Stroke Center Hospital. The binary logistic regression model was used to analyze the association between the types of stroke and some covariates namely age, sex, total cholesterol, blood sugar level, and history of diseases (hypertension/stroke/diabetes mellitus). Maximum Likelihood Estimation was used to estimate parameters. Combinations of covariates were compared using goodness-of-fit measures. Comparisons were made in the context of a case study, namely stroke patients (2017-2020). The results showed that a binary logistic model combining the history of diseases and blood sugar level provided the most suitable model as it has the smallest AIC and covariates included are statistically significant. The coefficient estimation of the history of diseases variable is -0.92402 with an odds ratio value exp(-0.92402)=0.4. This means that stroke patients who have a history of diseases experience a reduction of 60% in the odds of having a hemorrhagic stroke compared to stroke patients that do not have a history of diseases. In other words, stroke patients who have a history of diseases tend to have a non-hemorrhagic stroke. Furthermore, the coefficient estimation of blood sugar level is 0.74395 with an odds ratio value exp(0.74395)=2. It means that stroke patients who do not have normal blood sugar levels tend to have a hemorrhagic stroke 2 times greater than stroke patients with normal blood sugar levels. A history of diseases and blood sugar level were factors that significantly affect the types of stroke.
Application of the Naive Bayes Algorithm for Classification of Family Hope Program (FHP) Assistance Recipients Nunung Marlika; Aswi; Suwardi Annas
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 8 No. 01 (2026)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm418

Abstract

One of the classification methods commonly used to determine the eligibility of recipients of the Familiy Hope Program (FHP) is the Naive Bayes Algorithm, also often called the Naive Bayes Classifier. This method is probability to classify data quickly and efficiently for eligibity analysis in social assistance programs. Naive Bayes is classification that uses a probabiity and statistical approach to group data. In this study, the Naive Bayes Algorithm was applied in classifying recipients of the Family Hope Program and to determine the level of accuracy, recall and precision of the Naive Bayes method. The results of this study are the accuracy, recall and precision of the Naive Bayes methode. The results of this study are the accuracy value produced by the Naive Bayes methode of 90% in the division of training and testing data 60%:40% the accurqacy value of 93% in the division of training and testing data 70%:30%, and the accuracy values of 90% ih the division of training and testing data 80%:20%.
Application of the Mixed Geographically Weighted Regression Model to Identify Influencing Factors for Literacy Development Index of Indonesian Society's in 2022 Zulhijrah Zulhijrah; Ruliana Ruliana; Aswi Aswi
Indonesian Journal of Applied Statistics Vol 7, No 2 (2024)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v7i2.88784

Abstract

The mixed geographically weighted regression (MGWR) method is a combination of a linear regression model and a geographically weighted regression (GWR) model. The MGWR model can produce parameter estimates that have global parameter estimates, and other parameters that have local parameters according to the observation location. This method can be used in epidemiological studies that are influenced by spatial heterogeneity. The aim of this research is to determine and model the factors that influence the Community Literacy Development Index (CLDI) in Indonesia based on MGWR modeling. The data used in this research is CLDI data in Indonesia in 2022 along with the factors that are thought to influence it. The results of this research indicate that the MGWR model outperforms both the linear regression and GWR models, as it yields the lowest Akaike information criterion (AIC) value and an ?² value of 96.54%. Based on the modeling results, several factors influencing CLDI were identified, including the percentage of libraries, the adequacy ratio of library collections, the average length of schooling, and the level of participation in organized learning. Keywords: Literacy; literacy development index; mixed geographically weighted regression; spatial
Co-Authors A. Nurul Amalia AA Sudharmawan, AA Abdul Rahman Abdul Rahmat Abidin, Muh. Zulkifli Abidin, Muhammad Rais Ahmar, Ansari Saleh Ambo Upe Andi Feriansyah Andi Feriansyah Andi Gagah Palarung Taufik Andi Gagah Palarungi Taufik Andi Gagah Palarungi Taufik Andi Ilham Azhar Mangkona Andi Muhammad Ridho Yusuf Sainon Andin P Andi Shahifah Muthahharah Annas, Suwardi Annas, Suwardi Annas, Suwardi Annas, Suwarni Arismunandar Arismunandar Asrirawan Assagaf, Said Fachry Awaluddin Awi Dassa Awi Dassa, Awi Awi, Awi Bakri, Nurul Aulya Besse Sulfiani Bobby Poerwanto Bobby Poerwanto Bobby Poerwanto BQ Nurkhalishah K. Bustan, Muhammad Nadjib Cramb, Susanna Derliani Natalia Dewi Aprilia Wardani Syam Diana Eka Pratiwi Ernawati Ernawati Fahmuddin, Muhammad Fahmuddin, Muhammad Fajar Arwadi Folorunso, Serifat Adedamola Haekal, Muh. Fahri Halimah Husain Hammado, Nurussyariah Hisyam Ihsan Idul Fitri Abdullah Ikhwana, Nur Irwan Irwan Irwan, Irwan Ishma Azizah S Isnaini, Mardatunnisa Isnaini, Wulan Maulia Kasmir Kasmir, Kasmir La Ode Salman Yassar Lalu Ramzy Rahmanda M Nadjib Bustan M. Miftach Fakhri Mahadtir, Muhamad Mangkona, Andi Ilham Azhar Mar'ah, Zakiyah Mardatunnisa Isnaini Maria Dominggo Wea Mauliyana, Andi Meyrna Vidya Shanty Muhammad Abdy Muhammad Abdy Muhammad Abdy Muhammad Ammar Naufal Muhammad Arif Tiro Muhammad Arif Tiro Muhammad Arif Tiro Muhammad Arif Tiro Muhammad Arif Tiro Muhammad Arif Tiro, Muhammad Arif Muhammad Fahmuddin Muhammad Fahmuddin Muhammad Fahmuddin Sudding Muhammad Kasim Aidid Muhammad Kasim Aidid Muhammad Mahadtir Muttaqin, Imam Akbar Natalia, Derliani Nini Harnikayani Hasa Novianti, Andi Rima Nunung Marlika Nur Aziza S Nur Salsabila Wero Nurhikmawati, Nurhikmawati Nurhilaliyah Nurhilaliyah Nurhilaliyah Nurhilaliyah Nurhilaliyah Nurhilaliyah, Nurhilaliyah Nurkaila Kaito Nurlaila Kaito Nurlia Nurlia Nurul Aulya Bakri Nurul Fadilah Syahrul Nurul Ilmi Nusrang, Muhammad Oktaviana Oktaviana Oktaviana Oktaviana Palarungi, Andi Gagah Panessai Sir Poerwanto, Bobby Poerwanto, Bobby Poewanto, Bobby Putri Ananda, Elma Yulia Putri, Siti Choiratun Aisyah Putri, Siti Choirotun Aisyah Rahma, Ina Rahman, Abdul Rahmat Hidayat Rahmat Hidayat Rahmat Hidayat Rahmat Hidayat Rahmat Hidayat Rahmawati Rahmawati Rahmawati Ramadani, Reski Aulia Reski Aulia Ramadani Rezki Amalia Idrus Rika Saniarti Riska Saputri Risma Mastory Ruliana Ruliana Ruliana Ruliana Ruliana Ruliana Ruliana, Ruliana S, Muhammad Fahmuddin Sahlan Sidjara Saleh, Andi Rahmat Salsabila, Afifah Sapriani Shanty, Meyrna Vidya Siti Choirotun Aisyah Putri Siti Choirotun Aisyah Putri Sitti Aminah Sri Ayu Astuti Sri Rahayu Stevani Stevani Suardi, Shafira Suci Amaliah Sudarmin Sudarmin Sudarmin Sudarmin Sudarmin Sudarmin Sudarmin Sudarmin Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna Sukarna, Sukarna Sulistiawaty Sulistiawaty, Sulistiawaty Sumarni Sumarni Supriadi Yusuf Susanna Cramb Suwardi Anas Suwardi Annas Suwardi Annas Syafruddin Side Syamsiar, Syamsiar Taufik, Andi Gagah Palarungi Vivianti Vivianti Vivianti Wahidah Sanusi Wahyudin Wahyudin Wanda Yudi Yassar, La Ode Salman Yudi, Wanda Yunus, Sitti Rahma Zakiah, Andi Sitti Zakiyah Mar'ah Zulhijrah Zulhijrah Zulhijrah Zulhijrah Zulhijrah Zulkifli Rais