Impacts of Voice and Accountability upon Innovation in Pakistan: Evidence from ARDL and Maki Cointegration Approaches

Table 5

ARDL long- and short-run estimates with year dummy.

Variable	Coefficient	Standard error	T-statistic
Long-run estimates
Constant	11.212	22.627	0.495
VAA	−8.901	3.669	−2.425
ln TSM	−0.022	0.550	−0.041
ln INF	−4.372	3.674	−1.189
ln SAV	−0.675	2.982	−0.226
D1 2005	0.945	1.049	0.900
D2 2007	0.555	1.665	0.333
D3 2008	0.012	1.178	0.010
Short-run estimates
Δln INN (−1)	0.878	0.077	11.3137
VAA	−17.156	1.386	−12.3743
ΔVAA(−1)	16.218	1.623	9.987
ln TSM	−0.532	0.122	−4.340
Δln TSM(−1)	0.440	0.129	3.414
ln INF	−5.152	0.883	−5.833
Δln INF(−1)	5.627	0.951	5.915
ln SAV	−2.848	0.764	−3.727
Δln SAV(−1)	2.569	0.827	3.105
D1 2005	0.617	0.137	4.487
Δ D1 2005	−0.671	0.157	−4.260
D2 2007	−0.958	0.130	−7.360
Δ D2 2007	0.897	0.152	5.886
D3 2008	−0.497	0.132	3.754
Δ D3 2008	−0.606	0.142	−4.251
ECT(−1)	−0.138	0.023	−5.886
R-square	0.978
Adjusted R-square	0.963
F-statistic	65.173
Prob(F-statistic):	0.000
Lag length selection criteria	AIC
Bound testing cointegration analysis (F-statistic values)	3.12

Note. , , and Significant level at 10%, 5%, and 1%.