PAKPAHAN, POSMA CHRISTIAN FELIX (2026) PERAMALAN HARGA HARIAN BAWANG MERAH DI JAWA TENGAH MENGGUNAKAN MODEL N-HITS. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
cover.pdf Download (668kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
bab 1.pdf Restricted to Registered users only Download (370kB) |
|
|
Text (BAB II)
bab 2.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB III)
bab 3.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB IV)
bab 4.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB V)
bab 5.pdf Restricted to Registered users only Download (341kB) |
|
|
Text (DAFTAR PUSTAKA)
dafpus.pdf Restricted to Registered users only Download (304kB) |
|
|
Text (LAMPIRAN)
lampiran.pdf Restricted to Registered users only Download (617kB) |
Abstract
Fluctuations in shallot prices are one of the factors affecting food price stability in Indonesia, making accurate forecasting methods essential for supporting decisionmaking. This study aims to develop a daily shallot price forecasting model for Central Java Province using the Neural Hierarchical Interpolation for Time Series (N-HITS) model. The dataset consists of daily shallot price data from December 1, 2017, to December 1, 2025, obtained from the National Strategic Food Price Information Center (PIHPS), Bank Indonesia. The preprocessing stage included daily calendar reconstruction, missing value imputation using the Last Observation Carried Forward (LOCF) method, Min-Max normalization, and data splitting into training, validation, and testing sets. The N-HITS model was configured with an input window of 180 days and a forecast horizon of 30 days. Model performance was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). Experimental results show that the proposed model achieved an MAE of IDR 2,058.24, an RMSE of IDR 3,001.10, and a MAPE of 4.58% on the testing dataset. The obtained MAPE indicates that the model provides relatively accurate forecasts for daily shallot prices. Therefore, the N-HITS model demonstrates its potential as a deep learningbased forecasting approach for predicting agricultural commodity prices and can serve as a reference for developing forecasting systems to support decision-making in the food sector. Kata kunci: N-HITS Model for Time Series Forecasting of Shallot Prices in Central Java Using Deep Learning. Fluktuasi harga bawang merah merupakan salah satu permasalahan yang memengaruhi stabilitas harga pangan di Indonesia sehingga diperlukan metode peramalan yang mampu menghasilkan prediksi secara akurat. Penelitian ini bertujuan membangun model peramalan harga harian bawang merah di Provinsi Jawa Tengah menggunakan model Neural Hierarchical Interpolation for Time Series (N-HITS). Dataset yang digunakan merupakan data harga harian bawang merah periode 1 Desember 2017 hingga 1 Desember 2025 yang diperoleh dari Pusat Informasi Harga Pangan Strategis Nasional (PIHPS) Bank Indonesia. Tahap preprocessing meliputi rekonstruksi kalender harian, imputasi nilai hilang menggunakan Last Observation Carried Forward (LOCF), normalisasi Min-Max, serta pembagian data menjadi data latih, validasi, dan uji. Model N-HITS dikonfigurasi dengan input window selama 180 hari dan forecast horizon selama 30 hari. Kinerja model dievaluasi menggunakan Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), dan Mean Absolute Percentage Error (MAPE). Hasil pengujian menunjukkan bahwa model menghasilkan nilai MAE sebesar Rp2.058,24, RMSE sebesar Rp3.001,10, dan MAPE sebesar 4,58% pada data uji. Nilai MAPE tersebut menunjukkan bahwa model memiliki tingkat kesalahan prediksi yang rendah sehingga mampu memprediksi harga harian bawang merah dengan baik. Hasil penelitian ini diharapkan dapat menjadi referensi dalam pengembangan sistem peramalan harga komoditas pangan berbasis deep learning. Kata kunci: N-HITS, peramalan, bawang merah, deret waktu, deep learning.
Actions (login required)
 |
View Item |
