Table 5: Estimated Causal Models for TC and JTC

In logs

Excel | CSV

	∇12TC	∇∇12JTC
Constant	0.0544*	-0.0368*
	(0.0073)	(0.0076)
DAYS	0.0018	0.0026*
	(0.0024)	(0.0010)
EASTER	-0.0535*	-0.0710*
	(0.0106)	(0.0041)
FEB93		0.0579*
		(0.0136)
APR93		-0.0870*
		(0.0137)
MAR94	0.0368*
	(0.0165)
JUL95		0.0492*
		(0.0085)
ΔPTCi	0.1576	-0.0134
	(0.3032)	(0.1387)
MRLi	0.2517*B4 - 0.4175*B5	0.1800*B4 - 0.2087*B5
	(0.1344) (0.1665)	(0.0861) (0.0977)
	(1 + 0.9897*B)
	(0.0672)
θ1	0.4557*
	(0.1316)
θ2	-0.5399*	0.4671*
	(0.0997)	(0.1641)
θ3		-0.5261*
		(0.1628)
Θ12	-0.1285	-0.4048*
	(0.0939)	(0.0949)
φ1	0.3240*	0.9753*
	(0.1042)	(0.1044)
φ2		-0.6448*
		(0.1997)
φ3	0.2694*	0.3357*
	(0.0964)	(0.1314)
φ4		-0.3400*
		(0.0942)
	0.0256	0.03035
R2	0.7	0.86
LBQ (12, 24, 36)	11.2, 22.8, 45.6	10.6, 14.6, 21.9
Jarque-Bera (p-value)	2.75 (0.252)	2.69 (0.260)
White's heterosk (p-value)	6.68 (0.946)	5.86 (0.97)

Notes: Standard errors in parenthesis. * represents significant at 5%; ∇ = (1 - B) and ∇₁₂ = (1 - B¹²): regular and seasonal differences; B: lag operator; ΔPTC_i (i = TC, JTC): rates of growth of deflated price for TC tickets; MRL: Metro route length; LBQ: Ljung-Box Q statistics; = residual standard error.