부산대학교 도서관

This paper presents an original method for estimating a density. Theestimator is mixture of kernel estimators of components of an``approximate normal mixture model'' for the data. The formula of theAsymptotic Mean Integrated Square Error is established. A simulationis conducted to assess the performance of the method.\parAlthough the simulation carried out in the paper suggests that the newmethod might be superior under certain circumstances, the method needsfurther studies to be considered as a sound alternative to classicalapproaches. The consistency properties are proven when the parametersof the mixing model are nonrandom. However, in practice (as in thesimulation), they are to be estimated. Also, the selection of thenumber of mixture components is yet to be addressed.