Predictions of short-term traffic volume support proactive transportation management and traveler information services. To address the periodicity, nonlinearity, uncertainty, and complexity of short-term traffic forecasting, seasonal autoregressive integrated moving average (SARIMA) and support vector machine (SVM) models are often used separately to forecast the time series of traffic flow. SARIMA can discover intrinsic relations (correlations) in time series data, especially fit for modeling seasonal, stochastic time series. In contrast, SVM has a strong nonlinear mapping ability for input and output, appropriate for solving complex, nonlinear problems. The few combined schemes that have been presented previously tend to be complicated. A simple and effective novel hybrid methodology that combines the SARIMA and SVM models is proposed to take advantage of the individual strengths of the two models. Two key issues in building a hybrid model process are discussed: identification of SVM input dimensions via SARIMA and parameter optimization of the hybrid model with particle swarm optimization. Experimental results with real-world data sets indicated that the hybrid model was superior to the individual SARIMA or SVM model in terms of root mean square error and mean absolute percentage error for the prediction of the short-term traffic flow.