Forecasting Monthly Car Sales of Toyota and Daihatsu Using SARIMA Box-Jenkins Method

Authors

  • Richella Yasper President University image/svg+xml
  • Feriza Anggraeni
  • Edwin Nugraha

DOI:

https://doi.org/10.33021/88sfnf62

Abstract

Accurate sales forecasting is essential in the automotive industry to support production planning and strategic decision-making, particularly for vehicle sales data characterized by seasonal fluctuations. This study aims to forecast the monthly sales of Toyota and Daihatsu in Indonesia using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. The process includes stationarity testing using the Augmented Dickey–Fuller (ADF) test, model identification through ACF and PACF plots, parameter estimation, residual diagnostic checking using the Shapiro–Wilk and Ljung–Box tests, model selection based on the lowest AIC and BIC, and error evaluation. Based on the analysis, SARIMA(2,1,1)(0,1,1)[12] was identified as the most appropriate model for Toyota sales forecasting, achieving a MAPE value of 14.91%. Meanwhile, SARIMA(0,1,1)(1,1,1)[12] was selected as the optimal model for Daihatsu sales forecasting with a MAPE value of 16.39%. The findings demonstrate that the SARIMA model is capable of ef ectively capturing seasonal patterns and generating reliable forecasting performance. Therefore, the proposed models can provide valuable insights for improving sales planning and supporting strategic decisions within the automotive industry.

Downloads

Published

2026-09-28

How to Cite

Forecasting Monthly Car Sales of Toyota and Daihatsu Using SARIMA Box-Jenkins Method. (2026). Journal of Actuarial, Finance, and Risk Management, 5(1). https://doi.org/10.33021/88sfnf62