Claim Missing Document
Check
Articles

Found 3 Documents
Search

Pemodelan Kurva Engel Non Makanan dengan Bayesian Quantile Regression di Provinsi Papua Muhammad Fajar; Setiawan
Statistika Vol. 22 No. 1 (2022): Statistika
Publisher : Department of Statistics, Faculty of Mathematics and Natural Sciences, Universitas Islam Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/statistika.v22i1.335

Abstract

ABSTRAK Tujuan paper ini adalah untuk melakukan pemodelan kurva Engel non makanan, dimana pada sebaran data mengindikasikan terjadinya heterokedastisitas. Metode yang digunakan dalam penelitian ini adalah regresi kuantil dengan pendekatan Bayesian. Sumber data yang digunakan adalah data pengeluaran konsumsi rumah tangga dan pengeluaran konsumsi rumah tangga untuk non makanan di Provinsi Papua, yang berasal dari SUSENAS Maret 2018. SUSENAS adalah salah satu survei yang dilaksanakan Badan Pusat Statistik. Rumah tangga sampel SUSENAS Maret 2018 sebanyak 10629 rumah tangga. Hasil dari penelitian ini adalah regresi kuantil Bayesian jauh lebih representatif karena mencakup sebaran data, baik pada sentral data maupun daerah disekitar sentral data, dibandingkan regresi linear biasa, dimana regresi linear biasa hanya mencakup berkisar pada sentral data. Ini berimplikasi bahwa pemodelan kurva Engel dengan regresi kuantil adalah sangat cocok dengan sebaran data. Pengeluaran konsumsi rumah tangga berpengaruh positif dan signifikan pada pengeluran konsumsi rumah tangga untuk non makanan. Proporsi pengeluaran konsumsi rumah tangga untuk non makanan berbeda-beda untuk setiap tingkat pengeluarannya. Semakin tinggi pengeluaran rumah tangga, maka proporsi konsumsi untuk non makanan, semakin besar proporsinya. Dan sebaliknya, rumah tangga dengan level pengeluaran rendah, Proporsi konsumsi untuk makan dan minum lebih besar daripada konsumsi untuk non makanan. ABSTRACT This paper aims to model a non-food Engel curve, where the distribution of data shows heteroscedasticity. The method used in this research is quantile regression with the Bayesian approach. The data sources used are household consumption expenditure data and household nonfood consumption expenditures in Papua Province which are derived from SUSENAS March 2018. SUSENAS is one of the surveys conducted by the Badan Pusat Statistik-Statistics Indonesia. SUSENAS March 2018 sample households were 10629 households. The result of this study is that Bayesian quantile regression is much more representative covering the distribution of data, both in the data center and around it, compared to ordinary linear regression, where ordinary linear regression only has a wide range because of the central data. This implies that the modeling of the Engel curve with quantile regression is very suitable for the distribution of the data. Household consumption expenditure has a positive and significant effect on non-food household consumption expenditure. The proportion of household consumption expenditure on food varies for each level of competition. The higher the household expenditure, the higher the proportion of non-food consumption. And conversely, in households with low levels of expenditure, the proportion of consumption for food is greater than consumption for non-food.
SAR-to-Optical Image Translation Based on CycleGAN with Training Stabilization: A Sumatra Flood Case Study Ahmad Imdad; Kartika Fithriasari; Setiawan
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/mathscience4878

Abstract

Flood monitoring in tropical regions is challenged by persistent cloud cover, which restricts optical imagery, while consistently available Synthetic Aperture Radar (SAR) imagery presents difficulties in visual interpretation. This study employs CycleGAN to translate SAR into optical-like imagery in an unpaired-data scenario within flood-affected areas of Sumatra. Seven training configurations were evaluated, including the default setting, an asymmetric learning rate scheme, a combination of spectral normalization and geometric augmentation, isolated ablations of these components, and variations in the cycle-consistency coefficient. The dataset consisted of 482 Sentinel-1 SAR patches and 446 Sentinel-2 optical patches for training, alongside 276 SAR images for testing, all acquired via Google Earth Engine during the November 2025 flood. The evaluation utilized four metrics: the Structural Similarity Index (SSIM), Learned Perceptual Image Patch Similarity (LPIPS), Fréchet Inception Distance (FID), and the convergence epoch as the primary model-selection criterion. SSIM and LPIPS were calculated for cyclic reconstruction to address the lack of paired optical references. The configuration integrating spectral normalization, a reduced discriminator learning rate, and geometric augmentation achieved the fastest convergence (epoch 43) and the highest performance across all metrics (SSIM = 0.939, LPIPS = 0.025, and FID = 143.00).
Identifikasi Permasalahan Mahasiswa Evaluasi Semester Program Akademik dan Vokasi di Institut Teknologi Sepuluh Nopember Surabaya Yuli Purwanto; Setiawan; Sunawi; Fatkhul Iman Nurdjati; Siti Machmudah
Jurnal Sosial Humaniora Vol 17 No 2 (2024)
Publisher : Direktorat Riset dan Pengabdian Kepada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j24433527.v17i2.21982

Abstract

This study investigates the issues faced by students with Dispensation, ProbationSemester Promotion, and Non-Promotion statuses based on evaluations from the2021 Even Semester to the 2024 Odd Semester at Institut Teknologi SepuluhNopember, Surabaya. Using a quantitative descriptive approach, the researchexamines problems in semester evaluations. The population consists of 422 activestudents, with a sample of 252 students selected through Proportioned StratifiedRandom Sampling. Data was collected via questionnaires and interviews.Analysis employed the Guidance and Counseling Problem Checklist (DCM) inHigher Education, covering 11 topics through 56 Likert-scale statements.Problems were ranked based on Respondent Achievement Level (TCR)percentages. Findings reveal that the most significant issue is in the "VeryProblematic" category. The top problem topic is Campus Adjustment (69.78%),with the leading issue being motivation to study (88.73%). The second issue isFuture and Ambition (55.94%), with a focus on identifying self-potential(79.60%). Social Life and Organizational Activity ranks third (52.56%),primarily due to lack of organizational participation (60.79%). Lastly, Recreationand Hobbies (51.40%) highlights insufficient time for hobbies to boostenthusiasm (65.56%). These findings underscore key areas for student supportinterventions