TableĀ 6

Panel asymmetric causality (Inf+ and pov+)

Cross sectionsNull hypothesis
Inf+ does not cause pov+Pov+ does not cause inf+
Wald statp-valuelagDecisionWald statp-valuelagDecision
Panel55.695***0.00611Reject3.5120.2201Accept
Botswana28.5060.15211Accept3.446*0.0721Reject
Burkina Faso82.095**0.01911Reject2.4230.1291Accept
Cameroon9.9050.59111Accept6.320**0.0171Reject
Congo, Rep104.645**0.01111Reject0.8230.3701Accept
Gabon40.183*0.08211Reject2.954*0.0941Reject
Kenya21.3150.24111Accept1.5630.2201Accept
Madagascar148.935***0.00511Reject1.7360.1961Accept
Mauritania19.2850.27811Accept4.931**0.0331Reject
Nigeria94.206**0.01411Reject0.0060.9381Accept
Senegal43.404*0.07111Reject0.8330.3681Accept
South Africa22.4350.22311Accept12.755***0.0011Reject
Togo53.429**0.04711Reject4.354**0.0441Reject

Note(s): The symbols ***, ** and * represent 1%, 5 and 10% level of significance, respectively. Also, computation of p-values are based on 1,000 bootstrap replications. The lag length criteria is endogenously determined based on Bayesian information criteria (BIC)

Source(s): Authors' computations

or Create an Account

Close Modal
Close Modal