Effects of Coverage Area Treatment, Spatial Analysis Unit, and Regression Model on the Results of Station-Level Demand Modeling of Urban Rail Transit

Table 7

Spatial error model outputs.

		Equal division		Naive method		Thiessen polygons
		CBG	CT	CBG	CT	CBG	CT
New York	Adjusted	0.567	0.565	0.438	0.449	0.477	0.557
San Francisco		0.771	0.763	0.763	0.764	0.768	0.788
Chicago		0.533	0.528	0.514	0.539	0.475	0.497
Philadelphia		0.897	0.903	0.886	0.920	0.898	0.928
Boston		0.663	0.595	0.509	0.474	0.581	0.547
New York	MAPE	74.733	75.294	89.749	91.908	95.056	85.800
San Francisco		43.068	42.702	42.300	42.100	45.948	43.021
Chicago		53.601	52.825	54.406	51.656	52.608	50.653
Philadelphia		138.200	111.300	141.637	117.328	201.967	152.672
Boston		106.656	126.006	183.444	208.402	151.789	173.288
New York	AIC	13897.300	13899.900	14006.800	13999.200	13976.300	13906.600
San Francisco		885.226	885.702	868.122	866.078	881.941	879.407
Chicago		4199.010	4201.250	4202.690	4196.470	4214.380	4209.230
Philadelphia		4041.238	4023.541	4072.510	4038.360	4072.340	4020.310
Boston		2499.250	2523.130	2550.900	2559.020	2527.610	2538.200