Booming with Speed: High-Speed Rail and Regional Green Innovation

Table 6

Heterogeneity analysis on urban development index and green sharing index.

	Panel A: urban development index			Panel B: green sharing index
	Low	Medium	High	Low	Medium	High
	GUMA			GUMA
	(1)	(2)	(3)	(1)	(2)	(3)
L.HSR	0.0191	0.0470	0.271	0.0702	0.119	0.264
	(0.0152)	(0.0232)	(0.0707)	(0.0325)	(0.0371)	(0.0652)
R^2	0.576	0.478	0.624	0.386	0.459	0.542
Test of group: coefficient difference	Medium vs. high			Low vs. medium	Medium vs. high	Low vs. high
Observed difference	−0.217			−0.049	−0.145	−0.194
	GIA			GIA
	(4)	(5)	(6)	(4)	(5)	(6)
L.HSR	0.00799	0.0641	0.318	0.135	0.125	0.289
	(0.0132)	(0.0249)	(0.0895)	(0.0478)	(0.0433)	(0.0853)
R^2	0.510	0.392	0.559	0.312	0.388	0.463
Test of group: coefficient difference	Medium vs. high			Low vs. medium	Medium vs. high	Low vs. high
Observed difference	−0.258			0.010	−0.165	−0.154
	GUMG			GUMG
	(7)	(8)	(9)	(7)	(8)	(9)
L.HSR	0.0164	0.0439	0.254	0.0621	0.105	0.254
	(0.0125)	(0.0183)	(0.0626)	(0.0306)	(0.0332)	(0.0590)
R^2	0.553	0.499	0.611	0.387	0.444	0.539
Test of group: coefficient difference	Medium vs. high			Low vs. medium	Medium vs. high	Low vs. high
Observed difference	−0.211			−0.043	−0.149	−0.192
	GIG			GIG
	(10)	(11)	(12)	(10)	(11)	(12)
L.HSR	0.00238	0.0147	0.0803	0.0231	0.0298	0.0818
	(0.00208)	(0.00487)	(0.0206)	(0.00915)	(0.00941)	(0.0212)
R^2	0.532	0.391	0.543	0.312	0.381	0.465
Test of group: coefficient difference	Medium vs. high			Low vs. medium	Medium vs. high	Low vs. high
Observed difference	−0.077			−0.007	−0.052	−0.059
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
City	Yes	Yes	Yes	Yes	Yes	Yes
N	1072	1104	1105	1105	1104	1072

Note: in Panel A, the estimation sample is divided into three groups based on cities’ urban development index (i.e., cities with low, medium, and high urban development index). In Panel B, the estimation sample is split based on cities’ green sharing index (i.e., cities with low, medium, and high green sharing index). The test of coefficient difference between groups is also carried out here using the bootstrap method. The bootstrap number in this paper is set to bs (1000). HSR and all control variables are lagged by one year. All regressions include city fixed effects and year fixed effects. Robust standard errors in parentheses are clustered at the city level. The symbols , , and indicate significance at the 1%, 5%, and 10% levels, respectively.