Table 6.

Results of the refined spatial error models with spatially lagged responses

	Model 1 (APE)		Model 2 (PE)
	Coefficient	VIF	Coefficient	VIF
Population growth rate	0.39***	2.04	−0.99***	2.16
Population size	−0.11***	2.32	0.04**	2.63
Young	/		0.04**	2.28
Old	/		−0.22***	2.94
Black	0.09***	1.58	/
College	0.06*	1.88	0.05*	2.04
High school	0.10**	4.62	/
Bachelor’s degree	0.06*	3.66	0.09**	2.75
Employment	−0.05*	2.72	/
Income	/		−0.06**	1.69
Agriculture	0.19***	3.41	−0.21***	4.04
Retail	/		0.04*	1.51
Public transportation	/		−0.07***	1.51
Commuting time	−0.11***	1.74	−0.16***	2.25
Airport accessibility	0.05*	1.24	/
Land developability	−0.11***	1.71	/
Neighbor average
Population growth rate	−0.19***	2.29	0.57***	3.66
Young	/		−0.04*	2.72
Old	/		0.14***	3.93
Hispanic	0.04	1.69	/
High school	−0.07*	3.28	/
Bachelor’s degree	/		−0.10***	3.31
Employment	/		−0.06***	2.36
Agriculture	−0.05*	1.94	/
Income	−0.06*	3.91	0.21***	4.47
Commuting time	/		0.07**	2.93
Airport accessibility	/		0.03**	1.67
Spatially lagged APE	0.18***	1.28	/
Spatially lagged PE	/		0.43***	1.87
Measures of fit
Adjusted R2	0.19		0.64
AIC	−8842.80		−9288.47

Notes:

APE = Absolute percentage error; PE = Percentage error; VIF = Variance inflation factor; AIC = Akaike information criterion.

Coefficients reported in the table are standardized coefficients.

“/” means that the corresponding independent variable is not included in the regression model.

^***p ≤ 0.001;

^**p ≤ 0.01;

^*p ≤0.05.

Adjusted R² penalizes the model for including too many predictors. It is computed using the formula $R_{\mathit{adj}}^2=1-[\frac{\left(1-R^2\right)(n-1)}{n-k-1}]$, where n is the sample size and k is the number of predictors.

The neighbor averages are calculated based on the first-order queen’s continuity matrix.