Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Journal of Informatics Management and Information Technology

Peramalan Harga Minyak Mentah Indonesia dengan Model Hibrida ARIMA–FTS Cheng Kusuma, Erin Jihan Wahyu; Zukhronah, Etik; Susanti, Yuliana
Journal of Informatics Management and Information Technology Vol. 5 No. 3 (2025): July 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/jimat.v5i3.629

Abstract

Economic growth is a key indicator of successful economic activities, with adequate crude oil availability playing a crucial role in supporting a country's economic development. This study aims to forecast Indonesian crude oil prices using an Autoregressive Integrated Moving Average (ARIMA)–Fuzzy Time Series (FTS) Cheng hybrid model. The data utilized consists of monthly Indonesian crude oil prices from January 2013 to April 2023 for training and from May 2023 to December 2024 for testing. The training data is modeled using ARIMA, and the residuals from the ARIMA model are subsequently analyzed using the FTS Cheng approach. The hybrid ARIMA-FTS Cheng forecast is generated by combining the predictions from both the ARIMA and FTS Cheng models. The results of the study show that the hybrid ARIMA–FTS Cheng model produced an MAPE of 7.46% on the training data and 4.57% on the testing data. Therefore, the ARIMA–FTS Cheng hybrid model is considered suitable for forecasting Indonesia's crude oil prices.
Model Hibrida ARIMA-Neural Network untuk Peramalan Kasus Tuberkulosis Agung Setyabudi, Arriza; Etik Zukhronah; Isnandar Slamet
Journal of Informatics Management and Information Technology Vol. 5 No. 3 (2025): July 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/jimat.v5i3.597

Abstract

Tuberculosis (TB) remains a significant public health challenge in Surakarta City, necessitating accurate forecasting methods to support effective and planned control strategies. This study aims to evaluate the performance of the Autoregressive Integrated Moving Average-Neural Network (ARIMA-NN) hybrid model in forecasting monthly TB cases in the Surakarta region. The performance of this hybrid model is further compared with the ARIMA model. The research data used consists of monthly TB case data from January 2019 to September 2024 obtained from the Surakarta City Health Department. The data is divided into two sets: training data from January 2019 to December 2023 and testing data from January 2024 to September 2024. The ARIMA(0,1,1) model was identified as the best model for capturing the linear component of the data, yielding a Mean Absolute Percentage Error (MAPE) of 14.52% on the training data and 16.55% on the testing data. The residuals from the ARIMA(0,1,1) model were then further modeled using a Neural Network with 5 hidden neuron architecture, period lookback 6, and a learning rate of 0.1, to capture the remaining non-linear patterns. The developed ARIMA(0,1,1)-NN hybrid model showed better forecasting performance, with a MAPE value of 14.34% on the training data and 14.48% on the testing data. These results indicate that the ARIMA-NN hybrid approach offers the potential for improved accuracy compared to the ARIMA model in the context of TB case forecasting in Surakarta.