{"id":521,"date":"2017-09-24T06:37:52","date_gmt":"2017-09-24T06:37:52","guid":{"rendered":"http:\/\/jsr.isrt.ac.bd\/?post_type=article&p=521"},"modified":"2017-09-24T06:38:25","modified_gmt":"2017-09-24T06:38:25","slug":"gravitational-approach-density-estimation","status":"publish","type":"article","link":"http:\/\/jsr.isrt.ac.bd\/article\/gravitational-approach-density-estimation\/","title":{"rendered":"A gravitational approach to density estimation"},"content":{"rendered":"
It is shown that one can obtain a certain kernel density estimate at a given
\npoint by assuming that the sample points and are subject to the law of
\nuniversal gravitation or some variant thereof. Some distributional properties of
\nthe resulting kernel are discussed, including the asymptotic normality of a certain
\nrescaled kernel. A two-stage algorithm for the selection of an optimal bandwidth
\nis described. The proposed density estimation technique is applied to three widely
\nused data sets.<\/p>\n