A Hierarchical Passenger Mobility Prediction Model Applicable to Large Crowding Events

Table 5

The number of event participants from the main passenger sources of crowding station from 6:00 p.m. to 7:00 p.m.

Passenger source	Empirical data	MC model	-gram model	-gram model	-gram model
Shenzhen University	873	129	276	504	421
Qiaocheng North	604	135	522	694	561
Keyuan	563	198	782	1,071	1,120
Chegongmiao	559	131	501	503	673
High tech park	545	78	481	798	871
Huaqiang Road	535	89	331	428	508
Grand Theater	503	113	434	641	712
Shopping park	379	71	605	717	729
Huaqiang North	368	7	223	280	255
Laojie	361	46	64	198	223
MAPE (%)	/	97.7	68.8	60.2	64.9
MAE	/	125.6	74.0	60.3	67.8
RMSE	/	195.2	116.2	95.0	112.2