The consistency of the local kernel density estimator is proved. This nonparametric estimator is distinguished by its use of scaling matrices which are random and which may vary for each sample point. Its applications include adaptive construction of importance sampling functions.

Universities around the world are increasingly focusing on entrepreneurial activities (Slaughter and Lesley 1997; Clark 1998). In Australia, the growth of international entrepreneurial activities has resulted in the creation of a significant export-oriented sector. These activities include the...