Evaluation of ARIMA model performance in projecting future sales: case study on electronic products

Saputra, Bagus Hendra

Jurnal Teknik Informatika C.I.T. Medicom
Vol 16 No 5 (2024): November : Intelligent Decision Support System (IDSS)

The sales performance of electronic products is significantly affected by a variety of internal and external factors, necessitating precise forecasting models to aid strategic decision-making. This research investigates the effectiveness of ARIMA models in predicting future sales, focusing on a case study involving electronic products. The study utilizes monthly sales data obtained from company records and industry databases. The methodology includes assessing data stationarity through the Augmented Dickey-Fuller (ADF) test, applying differencing when required, and determining ARIMA parameters using Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) analyses. The findings reveal that ARIMA models effectively capture seasonal variations and trend patterns. Their performance is assessed using metrics such as Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). This study highlights the need to incorporate external factors into prediction models to enhance accuracy and recommends exploring alternative approaches that can better adapt to dynamic market conditions.