A Hybrid Model for Short-Term Traffic Flow Prediction Based on Variational Mode Decomposition, Wavelet Threshold Denoising, and Long Short-Term Memory Neural Network

Table 2

The prediction errors of different models using the traffic flow data of detector A.

Prediction models	RMSE (veh/5 min)	MAE (veh/5 min)	MAPE (%)
LSTM	11.077	8.285	15.360
WD-LSTM	10.850	8.066	10.912
EMD-WD-LSTM	10.217	7.626	14.227
VMD-WD-ANN	21.092	15.498	20.030
EEMD-ANN	22.305	16.936	25.391
TSD-BiLSTM	16.448	11.877	17.247
ARIMA-LSTM	12.138	8.846	12.612
VMD-WD-LSTM	6.888	5.116	9.313

Bold indicates the best value of each evaluation index.