Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal JURNAL ILMIAH MATEMATIKA DAN TERAPAN
Purnama, D I
Unknown Affiliation
Mathematics

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Peramalan Jumlah Penumpang Datang Melalui Transportasi Udara Di Sulawesi Tengah Menggunakan Support Vector Regression (SVR) Purnama, D I
JURNAL ILMIAH MATEMATIKA DAN TERAPAN Vol. 17 No. 1 (2020)
Publisher : Program Studi Matematika, Universitas Tadulako

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (505.585 KB) | DOI: 10.22487/2540766X.2020.v17.i1.15186

Abstract

The number of air transportation passengers in Central Sulawesi shows an increase and decrease every month. For this reason, a forecasting method is needed to predict the number of air transportation passengers in the future. Because the pattern of data on the number of air transportation passengers in Central Sulawesi Province has a nonlinear data pattern, a forecasting method is needed that can overcome these problems where in this study using the SVR model. In this study, the SVR model uses the RBF kernel function to overcome nonlinear data patterns and uses the grid search method to obtain the optimal parameters of the model.
Model Regresi Hurdle Negative Binomial (HNB) untuk Pemodelan Konsumsi Rokok di Provinsi Sulawesi Tengah Purnama, D I
JURNAL ILMIAH MATEMATIKA DAN TERAPAN Vol. 18 No. 1 (2021)
Publisher : Program Studi Matematika, Universitas Tadulako

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22487/2540766X.2021.v18.i1.15506

Abstract

The average expenditure on cigarettes per capita in Sulawesi Tengah Province has increased in 2020. There are several factors that can affect a person's cigarette consumption including gender, age, education and health. To model cigarette consumption with several influencing factors can be use the poison regression model or the Zero Inflated Poisson (ZIP) model. However, the two regression models cannot solve the excess zero and overdispersion problems so use the Hurdle Negative Binomial (HNB) regression model. The results of the analysis of cigarette consumption data in Central Sulawesi Province using the HNB model provide the best modeling results compared to the poisson regression model and the ZIP model because it has the smallest Akaike's Information Criterion (AIC) value. The results of testing the factors that significantly influence cigarette consumption in Central Sulawesi Province in the HNB regression model, namely the count model are gender, age and health. Whereas in the zerohurdle model, it is gender, age and education.
Peramalan Curah Hujan Di Kabupaten Parigi Moutong Menggunakan Model Seasonal Autoregressive Integrated Moving Average (SARIMA) Purnama, D I
JURNAL ILMIAH MATEMATIKA DAN TERAPAN Vol. 18 No. 2 (2021)
Publisher : Program Studi Matematika, Universitas Tadulako

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22487/2540766X.2021.v18.i2.15652

Abstract

Hujan merupakan fenomena alam yang sangat penting bagi kehidupan manusia. Hal ini membuat peramalan jumlah curah hujan di suatu daerah menjadi penting karena mampu mendukung proses pengambilan keputusan dalam berbagai sektor kehidupan. Dilain sisi perubahan iklim dunia membuat curah hujan seringkali susah diprediksi. Sehingga diperlukan identifikasi pola musiman pada data curah hujan sehingga mendukung peramalan curah hujan di suatu daerah. Tujuan dari penelitian ini adalah mengidentifkasi pola musiman serta menentukan model yang baik digunakan untuk meramalkan curah hujan di Kabupaten Parigi Moutong. Hasil identifikasi pola musiman mengggunakan regresi spektral menunjukkan bahwa data curah hujan di KabupatenParigi Moutong mengandung pola musiman. Selain itu diperoleh model Seasonal Autoregressive IntegratedMoving Average (ARIMA) terbaik untuk meramalkan curah hujan di Kabupaten Parigi Moutong adalah model SARIMA(1,1,0)(0,1,1)12. Model ini memiliki akurasi peramalan yang baik yang ditunjukkan dari nilai Mean Absolute Percentage Error (MAPE) sebesar 12,0157 pada data training dan 16,4647 pada data testing.
Co-Authors