Forecasting Electric Vehicle Sales Using Optimized SARIMA Model: A Two-Year Predictive Analysis

The rapid expansion of the electric vehicle (EV) market underscores the need for accurate forecasting models to guide decision-making for manufacturers, policymakers, and stakeholders. This study leverages the Seasonal Autoregressive Integrated Moving Average (SARIMA) model to predict monthly EV sales for the next two years based on historical sales data from January 2021 to December 2023. The data is sourced from the U.S. Department of Energy's 'Monthly Sales of New Light-Duty EVs in the United States' report. The SARIMA model is optimized through a comprehensive grid search, resulting in an optimal configuration of (1, 0, 2) for the non-seasonal component and (1, 0, 1, 12) for the seasonal component. The methodology involves preprocessing the sales data to confirm stationarity using the Augmented Dickey-Fuller (ADF) test. A grid search identifies the optimal parameters, with model performance evaluated using the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Hannan-Quinn Information Criterion (HQIC). The chosen model exhibits an AIC of 739.51, BIC of 749.01, and HQIC of 742.82, indicating a good fit. The forecasting results reveal a consistent upward trend in EV sales over the next 24 months, with the model predicting sales to reach approximately 96,076 units by January 2024, peaking at 108,559 units in July 2024 and slightly tapering off to 100,676 units by December 2025. These projections underscore the increasing consumer adoption of electric vehicles and provide valuable insights for industry stakeholders. With its consistent upward trend, the predicted growth trajectory highlights the potential for continued market expansion, driven by advancements in EV technology, increasing environmental awareness, and supportive governmental policies. In conclusion, the SARIMA model provides a reliable forecast of EV sales, facilitating informed strategic planning and resource allocation for industry participants. This research contributes to the broader understanding of market dynamics in the rapidly evolving electric vehicle sector. It underscores the importance of robust predictive analytics in supporting sustainable growth, instilling a sense of optimism and hope for the industry's future.