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

Table 4

Ordinary least squares regression model outputs.

		Equal division		Naive method		Thiessen polygons
		CBG	CT	CBG	CT	CBG	CT
New York	Adjusted	0.549	0.517	0.422	0.432	0.462	0.542
San Francisco		0.617	0.591	0.733	0.745	0.719	0.706
Chicago		0.445	0.428	0.431	0.443	0.369	0.382
Philadelphia		0.891	0.919	0.887	0.912	0.886	0.921
Boston		0.634	0.57	0.424	0.409	0.543	0.513
New York	MAPE	72.739	79.516	89.695	92.069	98.572	81.518
San Francisco		50.642	44.164	42.775	42.226	49.094	47.054
Chicago		59.19	57.231	62.261	60.258	64.85	63.097
Philadelphia		177.094	155.92	185.698	164.094	188.526	161.804
Boston		148.806	157.567	233.008	265.39	158.748	159.404
New York	AIC	13896.030	13923.020	14000.460	13993.310	13972.550	13903.600
San Francisco		885.742	886.852	868.122	866.078	872.189	874.187
Chicago		4201.277	4204.414	4203.672	4200.771	4216.351	4212.398
Philadelphia		4062.948	4019.140	4068.485	4032.550	4070.100	4015.053
Boston		2491.448	2511.762	2551.117	2554.565	2521.508	2528.015