Governance: A Source to Increase Tax Revenue in Pakistan
Table 5
ARDL bounds testing approach.
Estimated models
lnTAXREVt = f (lnCPIt, lnINDVAt, GSTABt)
lnTAXREVt = f (lnCPIt, lnINDVAt, LOAt)
lnTAXREVt = f (lnCPIt, lnINDVAt, ICONFt)
lnTAXREVt = f (lnCPIt, lnINDVAt, ECONFt)
Optimal lags
(1, 0, 2, 0)
(1, 0, 2, 0)
(1, 0, 0, 0)
(1, 0, 0, 0)
F-statistics
7.2945∗∗
6.1987∗∗
6.6014∗∗
5.9456∗∗
W-statistics
29.1779∗∗
24.7946∗∗
26.4055∗∗
23.7824∗∗
Significance level
Critical bounds for F-statistics
Critical bounds for W-statistics
Lower critical bound
Upper critical bound
Lower critical bound
Upper critical bound
5 per cent
3.4486
4.7653
13.7942
19.0614
10 per cent
2.8625
4.0255
11.4500
16.1019
Diagnostic tests
Serial correlation
0.3964 [0.529]
0.8371 [0.360]
1.7832 [0.182]
2.4380 [0.118]
Functional form
0.5551 [0.456]
1.0603 [0.303]
0.0013 [0.971]
0.7241 [0.395]
Normality
3.2374 [0.198]
0.5035 [0.777]
0.4867 [0.784]
0.5234 [0.770]
Heteroscedasticity
1.9956 [0.158]
2.6912 [0.101]
2.4543 [0.117]
2.4948 [0.114]
Note. Symbols “∗” and “∗∗” demonstrate the significance level at 10% and 5%, respectively. The values within square brackets represent probability values.
