Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Aiti: Jurnal Teknologi Informasi
Caesar, Yulius Bagus
Unknown Affiliation
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Evaluasi empiris model ARIMA dan LSTM dalam konteks peramalan penjualan mobil Toyota Caesar, Yulius Bagus; Hadiono, Kristophorus; Ardhianto, Eka
AITI Vol 22 No 2 (2025)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v22i2.221-235

Abstract

The increasing population and economic growth in major Indonesian cities drive demand for motor vehicles, particularly those of the Toyota brand. However, the fluctuating sales of Toyota from 2011 to 2023 complicate sales planning. This study aims to compare the performance of ARIMA and LSTM forecasting models in predicting Toyota car sales in Indonesia. The data used were obtained from the Association of Indonesian Automotive Industries (GAIKINDO). The results show that the LSTM model performs better than the ARIMA model in forecasting sales. The LSTM model yields lower prediction error values, as indicated by an RMSE of 5,198.40 and a MAPE of 15%, compared to ARIMA, which has an RMSE of 7,769.82 and a MAPE of 16%. Although the MAE of ARIMA is slightly better, at 4501.91, LSTM can minimize large prediction errors, which is evidenced by the significantly lower RMSE compared to ARIMA. The 1% difference in MAPE indicates that LSTM has a smaller percentage of prediction errors. These findings provide important implications for automotive companies in formulating more effective sales strategies. The LSTM model can be a valuable tool for anticipating market trends and making more accurate business decisions.
Co-Authors Eka Ardhianto Kristophorus Hadiono