Claim Missing Document
Check
Articles

Found 13 Documents
Search

Nonparametric Fourier Series Regression for Unemployment Analysis in Banten Province Bunga Miftahul Barokah; Fadhilah Fitri; Chairina Wirdiastuti
Rangkiang Mathematics Journal Vol. 5 No. 1 (2026): Rangkiang Mathematics Journal
Publisher : Department of Mathematics, Universitas Negeri Padang (UNP)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/rmj.v5i1.90

Abstract

The Open Unemployment Rate (OUR) is a vital indicator of regional economic performance, particularly in Banten Province, which faces disparities in education and poverty. This study models the unemployment rate using two predictors: average years of schooling and poverty level, through a nonparametric Fourier series regression for the 2017–2024 period. This method provides greater flexibility in capturing the nonlinear and fluctuating patterns often observed in socio-economic data. The analysis used secondary data from Statistics Indonesia (BPS), beginning with descriptive statistics and data visualization. Models were evaluated using Generalized Cross-Validation (GCV) and the coefficient of determination (R²). The optimal model was found at K = 3, with a GCV of 2.4057 and an R² of 0.5155. The model effectively captured the non-linear relationships between unemployment, education, and poverty. Although the R² value is moderate, this indicates that including additional explanatory variables could enhance the model’s performance. These findings support the use of Fourier series regression as an alternative approach for labor market analysis, especially when linear methods fall short and provide insights for developing more targeted employment policies.
Application of Algorithm Learning Vector Quantization for Air Quality Classification Roufsaldiaz Nawfal; Dina Fitria; Chairina Wirdiastuti
Mathematical Journal of Modelling and Forecasting Vol. 3 No. 2 (2025): December 2025
Publisher : Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/mjmf.v3i2.48

Abstract

This study aims to classify air quality using the Learning Vector Quantization (LVQ) algorithm based on the Air Quality and Pollution Assessment dataset obtained from Kaggle. The dataset comprises 5,000 observations, of which 4,000 were used for training and 1,000 for testing. The analytical process includes data preprocessing (normalization), the construction and training of the LVQ model, and performance evaluation using a confusion matrix. The experimental results demonstrate that the LVQ model successfully classified 903 of 1,000 test samples, yielding an overall accuracy of 90.3%. This level of accuracy indicates that the LVQ algorithm can capture relevant patterns in air quality variables and perform reliable classification across different air quality categories. The findings suggest that LVQ can serve as a potential foundation for developing automated air quality monitoring and decision-support systems. Future studies are encouraged to compare LVQ with other machine learning classification techniques to build a more optimal model and to gain deeper analytical insights.
Analisis Faktor-Faktor yang Mempengaruhi Prevalence of Undernourishment (PoU) di Indonesia Menggunakan Model Spatial Autoregressive (SAR) Rahmatul Annisa; Ervi Dayana Putri; Chairina Wirdiastuti
Emasains : Jurnal Edukasi Matematika dan Sains Vol. 15 No. 2 (2026): Juli 2026
Publisher : Program Studi Pendidikan Matematika dan Pendidikan Biologi Universitas PGRI Mahadewa Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59672/emasains.v15i2.6191

Abstract

Tujuan Pembangunan Berkelanjutan (Sustainable Development Goals/SDGs) merupakan kerangka pembangunan global yang dirancang untuk mewujudkan pembangunan berkelanjutan hingga tahun 2030. Di Indonesia, tingkat kerawanan pangan nasional mencapai 8,27 persen pada tahun 2024, menurun sekitar 0,26 poin dibandingkan tahun 2023 yang sebesar 8,53 persen. Meskipun terjadi penurunan secara nasional, prevalensi kekurangan gizi di beberapa provinsi di Indonesia masih tergolong tinggi. Hasil penelitian menunjukkan bahwa model Spatial Autoregressive (SAR) memiliki kinerja yang lebih baik dibandingkan model Ordinary Least Squares (OLS) dan Spatial Error Model (SEM). Dengan demikian, model SAR merupakan model terbaik dalam menganalisis faktor-faktor yang memengaruhi prevalensi kekurangan gizi antar provinsi di Indonesia tahun 2024. Hasil uji signifikansi model menunjukkan bahwa pendapatan per kapita, persentase rumah tangga dengan sanitasi layak, serta persentase rumah tangga dengan akses air bersih merupakan faktor yang berpengaruh terhadap kerawanan pangan di Indonesia. Nilai koefisien determinasi yang dihasilkan sebesar 75,56%, yang menunjukkan bahwa variabel independen dalam model mampu menjelaskan variasi variabel dependen secara cukup kuat.