Table 4

Results of the panel VAR model estimates

VariablesCO2RECInternetMobilegdpInst
CO20.433***−0.10454***3.368***27.70***−0.00922−0.944***
(0.111)(0.0001)(0.983)(4.617)(0.0114)(0.250)
REC−2.765***0.663***4.389−0.956***−8.8670***−3.016**
(0.00067)(0.00032)(0.000545)(0.000436)(0.0078)(0.0012)
Internet−0.00431*0.00145**0.804***0.468**0.0006440.0125
(0.00242)(0.0405)(0.0491)(0.182)(0.000978)(0.00928)
Mobile−0.000219−0.002390.0491***0.723***0.000514−0.00634**
(0.000792)(0.595)(0.00970)(0.0635)(0.000325)(0.00320)
gdp−0.406***0.1086***−14.25***−69.82***−0.364***5.459***
(0.0463)(0.0001)(1.138)(5.495)(0.0292)(0.450)
Inst0.0508**−0.0099*−0.2417.480**0.0471**0.362**
(0.0219)(0.0625)(0.156)(3.385)(0.0215)(0.173)
Observations817817817817817817

Note(s): The numbers in parentheses are the standard errors: ***p < 0.01, **p < 0.05 and * p < 0.1. The PVAR model is estimated by a fixed-effect GMM, taking into account the shifts leads to loss of data on the temporal dimension. The number of delays equal to 1 for the PVAR and 2 for the instruments; the number of instruments is equal to 1. To provide proof of the validity of the instruments, the specification test relating to the over-identification restrictions of Hansen (1982) was retained. Acceptance of the null hypothesis suggests the validity of the overidentifying restrictions. The results listed in the table show that the null hypothesis of the validity of the instruments is accepted at the 5% level

Source(s): Authors

[ViewLarge]
This Feature Is Available To Subscribers Only

or Create an Account

Close subscription notice
Close access options