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

Table 6

Generalized additive regression model outputs.

		Equal division		Naive method		Thiessen polygons
		CBG	CT	CBG	CT	CBG	CT
New York	Adjusted	0.683	0.706	0.539	0.529	0.689	0.653
San Francisco		0.98	0.984	0.971	0.974	0.965	0.967
Chicago		0.572	0.587	0.733	0.734	0.655	0.75
Philadelphia		0.982	0.987	0.99	0.989	0.987	0.987
Boston		0.635	0.66	0.456	0.427	0.559	0.647
New York	MAPE	69.19	64.682	79.478	84.718	78.939	85.246
San Francisco		12.224	10.923	13.763	14.431	15.845	16.421
Chicago		43.985	48.746	39.846	39.057	51.182	45.021
Philadelphia		113.19	103.895	79.543	85.854	92.655	104.403
Boston		150.868	133.957	201.751	263.342	169.054	182.116
New York	AIC	13761.000	13732.330	13919.390	13921.050	13756.010	13807.900
San Francisco		764.308	753.966	779.888	776.057	787.817	784.301
Chicago		4174.682	4171.752	4118.913	4118.002	4148.942	4112.932
Philadelphia		3813.419	3761.131	3733.719	3742.744	3769.477	3766.686
Boston		2491.280	2490.923	2546.696	2551.598	2517.802	2501.020