Ito (2001) has made an interesting argument on policy making in Japanese local governments. However, his analysis is based on the estimates from the models that do not properly take into account the temporal dependence of regional law adoption. The purpose of this comment is to point out a troubling aspect of Ito's event history modeling, and to put the existence of potential problems caused by temporally dependent observations into perspective.
[Statistical Methods for Data with Long-Range Dependence]: Comment