Growth, bank credit, and inflation in Mexico: evidence from an ARDL-bounds testing approach

Latin American Economic Review

Table 4 Testing for long run cointegration: F statistic and t statistic (Dependent variable: GDP)

Lag	Model A		Model B
	(GOV, INV, XP, INF, PC, INF × PC, LIB)		(GOV, INV, XP, INF, FD, INF × FD, LIB)
	P = 1	P = 2	P = 1	P = 2
F statistic	24.912*	2.888	34.294*	2.9833
t statistic	−5.108*	−2.576	−9.423*	−2.660
Lag length selection criteria
AIC	−5.442	−5.826	−5.013	−5.472
SBC	−4.724	−5.066	−5.472	−5.504

Lag	Model C		Model D
	(GOV, INV, XP, INF, BA, INF × BA, LIB)		(GOV, INV, XP, INF, LL, INF × LL, LIB)
	P = 1	P = 2	P = 1	P = 2
F statistic	25.147*	2.925	24.2905*	2.379
t statistic	−8.304*	−2.900	−6.690*	−0.644
Lag length selection criteria
AIC	−5.523	−4.954	−5.323	−5.446
SBC	−4.805	−5.440	−4.606	−4.686

Critical value bounds of the F statistic with k = 7 with constant (k = 8 is not available): (3.644, 5.464), (2.676, 4.130), and (2.260, 3.534) at the 1, 5, and 10 % level of significance, respectively
Critical value bounds of the t statistic with k = 7 with constant: (−2.58, −5.07), (−1.95, −4.43), and (−1.62, −4.09), respectively
*, **, and *** statistically significant at 1, 5, and 10 %, respectively
P is the lag length