cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Hasih Pratiwi
Contact Email
hpratiwi@mipa.uns.ac.id
Phone
+6282134673512
Journal Mail Official
ijas@mipa.uns.ac.id
Editorial Address
Study Program of Statistics, Universitas Sebelas Maret, Surakarta 57126, Indonesia
Location
Kota surakarta,
Jawa tengah
INDONESIA
Indonesian Journal of Applied Statistics
Published by Universitas Sebelas Maret
ISSN : -     EISSN : 2621086X     DOI : https://doi.org/10.13057/ijas
Core Subject : Economy, Science, Agriculture, Social,
Agriculture, Biological Sciences & Forestry Computer Science & IT Earth & Planetary Sciences Economics, Econometrics & Finance Environmental Science
Indonesian Journal of Applied Statistics (IJAS) is a journal published by Study Program of Statistics, Universitas Sebelas Maret, Surakarta, Indonesia. This journal is published twice every year, in May and November. The editors receive scientific papers on the results of research, scientific studies, and problem solving research using statistical method. Received papers will be reviewed to assess the substance of the material feasibility and technical writing.
Arjuna Subject : Matematika - Statistik dan Probabilitas
Articles 8 Documents
 1 
Search results for , issue "Vol 8, No 1 (2025)" : 8 Documents clear
Pengelompokan Provinsi di Indonesia Berdasarkan Indikator Pekerjaan Layak dengan Menggunakan K-Medoids Clustering Edy Widodo; Shifa Qonita; Safira Feri Amalina; Sifa Nurul Aoliya
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.97996

Abstract

Pekerjaan layak memainkan peran krusial dalam mengurangi kemiskinan dan menjadi salah satu tujuan utama dalam pembangunan berkelanjutan. Studi ini dilakukan untuk mengkaji berbagai indikator yang berperan dalam menentukan pekerjaan layak di Indonesia dan mengklasifikasikan tingkat kemampuan kerja di 34 provinsi berdasarkan indikator yang terkait dengan sustainable development goals (SDGs), khususnya Tujuan 8 tentang pekerjaan layak dan pertumbuhan ekonomi. Data yang digunakan meliputi indikator-indikator seperti tingkat pengangguran terbuka, proporsi pekerja informal, persentase anak usia 10-17 tahun yang bekerja, perusahaan yang menerapkan norma K3, dan persentase pemuda usia 15–24 tahun yang saat ini tidak terdaftar dalam pendidikan, pekerjaan, atau pelatihan (NEET). Teknik analisis yang digunakan adalah metode principal component analysis (PCA) dan metode k-medoids clustering. Berdasarkan analisis PCA, diperoleh 2 faktor utama, yaitu faktor kesejahteraan dan ketenagakerjaan serta faktor pengangguran dan keselamatan kerja. Secara bersama-sama, kedua faktor ini menyumbang 73,9516% dari keseluruhan varians dalam data. Analisis cluster dilakukan dengan menggunakan 2 faktor utama ini. Berdasarkan analisis pengelompokan menggunakan metode k-medoids, dihasilkan 3 cluster. Cluster 1 yang terdiri dari 20 provinsi merupakan wilayah dengan indikator pekerjaan layak yang moderat, cluster 2 yang terdiri dari 5 provinsi merupakan wilayah dengan indikator pekerjaan layak yang tinggi, dan cluster 3 yang terdiri dari 9 provinsi merupakan wilayah dengan indikator pekerjaan layak yang tinggi.Kata kunci: Pekerjaan layak; sustainable development goals (SDGs); k-medoids; principal component analysis (PCA).Decent work plays a crucial role in reducing poverty and serves as one of the key objectives in sustainable development. This study was conducted to examine various indicators that play a role in determining decent work in Indonesia and to classify the level of workability in 34 provinces based on indicators related to Sustainable Development Goals (SDGs), particularly Goal 8 on decent work and economic growth. The data used includes indicators such as the open unemployment rate, proportion of informal workers, percentage of children aged 10-17 years who work, companies that apply OSH norms, and the percentage of youth aged 15–24 who are not currently enrolled in education, work, or training (NEET). The analysis techniques used are Principal Component Analysis (PCA) method and K-Medoids Clustering method. Based on PCA analysis, 2 main factors were obtained, namely the welfare and employment factor and the unemployment and job safety factor. Together, these two factors account for 73.9516% of the overall variance in the data. Cluster analysis was conducted using these 2 main factors. Based on clustering analysis using the K-Medoids method, 3 clusters were generated. Cluster 1 consisting of 20 provinces is a region with moderate decent work indicators, cluster 2 consisting of 5 provinces is a region with high decent work indicators, and cluster 3 consisting of 9 provinces is a region with high decent work indicators.Keywords: decent work; sustainable development goals (SDGs); k-medoids; principal component analysis (PCA).
Penerapan Teknik Soft Voting Ensemble pada Klasifikasi Rating Film Alina Selia Rizka; Virgania Sari
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.100904

Abstract

Pertumbuhan jumlah penonton film yang begitu pesat mendorong industri perfilman untuk terus berinovasi, sehingga menghasilkan beragam judul baru dengan genre dan karakteristik yang semakin bervariasi. Kondisi ini menyebabkan kompleksitas data yang tinggi, sehingga dibutuhkan metode klasifikasi yang efektif dan akurat untuk mengelompokkan rating film berdasarkan karakteristiknya. Penelitian ini bertujuan untuk meningkatkan kinerja klasifikasi rating film dengan menggunakan metode ensemble soft voting, yang menggabungkan tiga algoritma klasifikasi, yaitu k-nearest neighbor (KNN), decision tree (DT), dan support vector machine (SVM). Evaluasi dilakukan dengan membandingkan kinerja metode soft voting terhadap masing-masing metode individu berdasarkan metrik akurasi, presisi, sensitivitas, dan F1-score. Hasil penelitian menunjukkan bahwa metode soft voting memberikan kinerja klasifikasi yang lebih baik dibandingkan metode KNN, decision tree, dan SVM secara terpisah, dengan capaian akurasi sebesar 89,64%, presisi 85,63%, sensitivitas 89,64%, dan nilai F1-score sebesar 86,52%.Kata kunci: klasifikasi, ensemble learning, KNN, decision tree, SVMThe rapid growth in the number of movie viewers has driven the film industry to continuously innovate, resulting in a diverse range of new titles with increasingly varied genres and characteristics. This has led to significant data complexity, necessitating an effective and accurate classification method to categorize movie ratings based on their characteristics. This study aims to evaluate the performance of the Soft Voting ensemble method in classifying movie ratings. The classification results from soft voting are compared to those of individual models, namely k-nearest neighbor (KNN), decision tree (DT), and support vector machine (SVM). The evaluation was conducted by comparing the performance of the soft voting method against each individual method based on accuracy, precision, sensitivity, and F1-score metrics. The results showed that the soft voting method provided better classification performance than the KNN, decision tree, and SVM methods individually, with an accuracy of 89.64%, a precision of 85.63%, a sensitivity of 89.64%, and an F1-score of 86.52%.Keywords: classification, ensemble learning, KNN, decision tree, SVM
Comparison of Hard Clustering and Soft Clustering Methods in Grouping Regencies/Cities in West Java Province Based on Regional Vulnerability Indicators to the Impact of Hydrometeorological Disasters in 2021 Hanifah Vida Indrasari; Yuliagnis Transver Wijaya
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.80562

Abstract

Indonesia is an archipelagic country with a high incidence of hydrometeorological disasters and the number is increasing every year. One of the provinces in Indonesia with the highest number of hydrometeorological disasters is West Java Province, where 98.97 percent are hydrometeorological disasters. This is also supported by the characteristics of the area which is dominated by mountains, high rainfall, has 40 watersheds, and has six faults that are suspected to be still active so that it is vulnerable to hydrometeorological disasters. Research on regional vulnerability to hydrometeorological disasters can be carried out by grouping regions based on the same level of vulnerability using the clustering method. The purpose of this study was to group regencies or cities in West Java Province based on indicators of regional vulnerability to the impacts of hydrometeorological disasters in 2021. The clustering method used is hard clustering (single linkage, complete linkage, average linkage, ward's method and k-means) and soft clustering (Fuzzy C-Means). The most optimal method for grouping regencies or cities in West Java Province is the complete linkage method with a total of 4 clusters. The result is that all the resulting clusters are vulnerable to the characteristics of social vulnerability.Keywords: cluster analysis; hard clustering; natural disasters; regional vulnerability; soft clustering.
Modeling and Classification Multicollinear Variables using Multinomial Ridge Logistic Regression Aprroach Giatma Dwijuna Ahadi; Ismaini Zain; Santi Puteri Rahayu
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.85795

Abstract

Multinomial Logistic Regression is a method used to find relationships between nominal or multinomial response variables (Y) with one or more predictor variables. Logistics Regression is a classic method that is often used to solve classification problems. Assumptions on Logistics Regression are models containing multicollinearity. Ridge Logistic Estimator (RLE) is methods to solve multicollinearity cases in Logistic Regression. Wu & Asar proposed a new ridge value that can also reduce bias in parameter estimation. Therefore, this research will discuss about Multinomial Ridge Logistic and selection the best of ridge constant values. The performance test of the ridge value will be applied to the Iris Dataset in R software. The best criteria for improvement ridge constant value by looking at the smallest standard error. The calculation results show that the Wu-Asar approach is the best ridge constant and Wald individual test shows significant results. Based on the result, show that the Wu-Asar Ridge constant value on Multinomial Ridge Logistic Regression are very good performance in estimated smaller standar error. The classification for dataset shows high results with 98% global accuracy.Keywords: multinomial; ridge logistic regression; Wu-Asar; standard error; classification
Perbandingan Metode Dekomposisi Multiplikatif dan Metode Prophet dalam Meramalkan Nilai Tukar USD ke Rupiah Mohammad Dwitiar Nalole; Agusyarif Rezka Nuha; La Ode Nashar; Novianita Achmad; Asriadi Asriadi
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.96572

Abstract

Nilai tukar antara USD dan Rupiah memiliki dampak signifikan terhadap perekonomian Indonesia, sehingga peramalan yang akurat menjadi sangat penting dalam pengambilan keputusan ekonomi. Penelitian ini membandingkan metode Dekomposisi Multiplikatif dan metode Prophet dalam meramalkan nilai tukar USD terhadap Rupiah. Data nilai tukar bulanan dari tahun 2018 hingga 2023 digunakan, dengan 80% sebagai data pelatihan dan 20% sebagai data pengujian. Model akhir untuk metode Dekomposisi Multiplikatif menggabungkan komponen tren, musiman, dan siklus secara multiplikatif, sehingga memungkinkan representasi perilaku nilai tukar yang lebih rinci dari waktu ke waktu. Sebaliknya, metode Prophet menghasilkan model akhir yang menggabungkan komponen tren dan musiman secara aditif, serta mampu mengakomodasi perubahan tren secara dinamis melalui deteksi otomatis terhadap changepoints. Akurasi kedua metode dievaluasi menggunakan Mean Absolute Percentage Error (MAPE). Hasil menunjukkan bahwa metode Dekomposisi Multiplikatif memiliki nilai MAPE sebesar 1,21%, sedangkan metode Prophet memiliki nilai MAPE sebesar 5,43%. Temuan ini menunjukkan bahwa metode Dekomposisi Multiplikatif lebih akurat dalam meramalkan nilai tukar untuk periode yang diberikan, sehingga lebih sesuai untuk dataset ini yang menunjukkan pola musiman yang kuat.Kata kunci: Nilai Tukar; Peramalan; Dekomposisi Multiplikatif; Prophet; MAPE.The exchange rate between USD and Rupiah has a significant impact on Indonesia's economy, making accurate forecasting essential for economic decision-making. This study compares the Multiplicative Decomposition method and the Prophet method in forecasting the USD to Rupiah exchange rate. Monthly exchange rate data from 2018 to 2023 was used, with an 80% training set and a 20% test set. The final model for the Multiplicative Decomposition method combines the trend, seasonal, and cycle components multiplicatively, allowing for a detailed representation of the exchange rate's behavior over time. In contrast, the Prophet method produces a final model that incorporates trend and seasonal components additively, while also accommodating dynamic changes in the trend through automatic detection of changepoints.. The accuracy of both methods was evaluated using Mean Absolute Percentage Error (MAPE). Results show that the Multiplicative Decomposition method had a MAPE of 1.21%, while the Prophet method had a MAPE of 5.43%. These findings indicate that the Multiplicative Decomposition method is more accurate in forecasting the exchange rate for the given period, making it more suitable for this dataset, which exhibits strong seasonal patterns.Keywords: Exchange rate; forecasting; Multiplicative Decomposition; Prophet; MAPE.
Penalized Spline Semiparametric Regression for Bivariate Response in Modeling Macro Poverty Indicators Cinta Rizki Oktarina; Idhia Sriliana; Sigit Nugroho
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.94370

Abstract

Semiparametric spline regression has become an increasingly popular method for modeling data due to its flexibility and objectivity, especially as a parameter estimation method. Spline functions are highly effective in semiparametric regression because they offer unique statistical interpretations by segmenting each predictor variable in relation to the response variable. Bivariate semiparametric regression can be applied to data where observations tend to have disparities between regions, making it suitable for poverty data, particularly the poverty depth index and the poverty severity index. The objective of this research is to analyze the models of the poverty depth index and poverty severity index, as well as to perform segmentation and interpretation of these models. This study utilized observations from 60 districts/cities in the southern part of Sumatra. Several predictor variables were considered, including the percentage of households with a floor area of ≤19 m², labor force participation rate, and life expectancy as parametric components, while the nonparametric components included the average length of schooling and the percentage of households with tap water sources. The estimation methods used were penalized least squares and penalized weighted least squares, involving a full search algorithm for selecting the number and location of knots. The results of the study indicated that the penalized weighted least squares method was the best estimator, with an MSE value of 0.3122 and two knots for each predictor, yielding GCV values of 4.3604 and 4.0794.Keywords: semiparametric regression; bivariate response; poverty; knot; penalized weighted least square
Metode Geographically Weighted Logistic Regression untuk Memodelkan Kasus Kemiskinan di Indonesia Tahun 2022 Ane Nurahmi; Dwi Agustin Nuriani Sirodj
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.92564

Abstract

Geographically Weighted Logistic Regression (GWLR) merupakan pengembangan dari model regresi logistik yang dirancang untuk menganalisis data spasial dengan variabel dependen kategorik. Penelitian ini bertujuan untuk memodelkan kasus kemiskinan di Indonesia pada tahun 2022 menggunakan fungsi pembobot Adaptive Gaussian Kernel serta mengidentifikasi faktor-faktor yang memengaruhinya. Hal ini penting mengingat garis kemiskinan nasional Indonesia (Rp 535.547/kapita/bulan) masih berada di bawah standar Bank Dunia (Rp 962.130/kapita/bulan). Tingkat kemiskinan di Indonesia menunjukkan variasi yang tinggi antarwilayah, yang dipengaruhi oleh perbedaan kondisi geografis dan karakteristik sosial ekonomi setempat. Dengan demikian, hubungan antara variabel-variabel penentu kemiskinan bersifat lokal dan bervariasi secara spasial. Pendekatan GWLR lebih tepat digunakan dibandingkan regresi logistik klasik karena mampu mengakomodasi heterogenitas spasial melalui pembobotan geografis. Kategori provinsi miskin ditetapkan berdasarkan nilai Head Count Index sebagai variabel dependen. Variabel independen yang dianalisis meliputi Pengeluaran Per Kapita Disesuaikan, Tingkat Pengangguran Terbuka, dan Upah Minimum Provinsi. Melalui penggunaan fungsi pembobot Adaptive Gaussian Kernel, diperoleh 34 model GWLR. Hasil penelitian menunjukkan bahwa Upah Minimum Provinsi berpengaruh signifikan terhadap tingkat kemiskinan pada delapan provinsi: Jawa Tengah, DI Yogyakarta, Jawa Timur, Bali, Nusa Tenggara Barat, Kalimantan Tengah, Kalimantan Selatan, dan Kalimantan Timur.Kata kunci: adaptive gaussian; geographically weighted logistic regression; kemiskinanGeographically Weighted Logistic Regression (GWLR) is a development of Logistic Regression for spatial data with a categorical dependent variable. The research aims to model poverty cases in Indonesia in 2022 using the Adaptive Gaussian Kernel weighting function and the factors that influence it, considering that Indonesia's poverty line is IDR 535,547/capita/month lower than the World Bank standard, IDR 962,130/capita/month. Poverty levels in Indonesia vary between regions due to different contributing factors based on geographical and socioeconomic conditions. Therefore, the relationship between variables that determine poverty is local and varies spatially, making the Geographically Weighted Logistic Regression (GWLR) method more appropriate than logistic regression because it is able to capture differences in influence between regions through geographical weighting. The poor province category is based on the Head Count Index value as the dependent variable. The dependent variables are adjusted Per Capita Expenditure, Open Unemployment Rate, and Provincial Minimum Wage. By using the Adaptive Gaussian Kernel weighting function, 34 models were obtained. The Provincial Minimum Wage has a significant effect on poverty cases in Indonesia in 8 provinces, namely the Provinces of Central Java, DI Yogyakarta, East Java, Bali, West Nusa Tenggara, Central Kalimantan, South Kalimantan and East Kalimantan.Keywords: adaptive gaussian; geographically weighted logistic regression; poverty
Modeling East Java Province Poverty Cases Using Birespon Truncted Spline Regression Rizka Amalia Putri; Nindya Wulandari; Erlyne Nadhilah Widyaningrum; Morina A. Fathan; Nur Rezky Safitriani
Indonesian Journal of Applied Statistics Vol 8, No 1 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v8i1.100915

Abstract

An analytical method for determining the relationship between predictor and response variables is regression. For data that shows unidentified patterns, nonparametric regression is a suitable data analysis technique. A nonparametric regression technique is the truncated spline. Due to the widespread use of truncated spline with a single response variable, this study employs biresponse truncated spline, which uses two response variables to produce a better model than single-response modeling. The purpose of this study is to obtain the best model and to identify which variables influence the poverty case in East Java Province using biresponse truncated spline regression. The best knot points were chosen for this investigation using Generalized Cross Validation (GCV). With three knot points and a model goodness of fit () of 95.83%, GCV gives the best modeling results. Applying this model to the East Java Province case of poverty using data on the poverty depth index and the percentage of the population living in poverty in 2023 reveals that the Labor Force Participation Rate (TPAK), Average Years of Schooling (RLS), and Open Unemployment Rate (TPT) all have a significant effect.Keywords: biresponse truncated spline; nonparametric regression; poverty

Page 1 of 1 | Total Record : 8

 1 

Filter by Year

2025 2025


Reset
Filter By Issues
All Issue Vol 8, No 2 (2025) Vol 8, No 1 (2025) Vol 7, No 2 (2024) Vol 7, No 1 (2024) Vol 6, No 2 (2023) Vol 6, No 1 (2023) Vol 5, No 2 (2022) Vol 5, No 1 (2022) Vol 4, No 2 (2021) Vol 4, No 1 (2021) Vol 3, No 2 (2020) Vol 3, No 1 (2020) Vol 2, No 2 (2019) Vol 2, No 1 (2019) Vol 1, No 2 (2018) Vol 1, No 1 (2018) More Issue