ON A ROLE OF A SCALE PARAMETER WHEN USING THE GAUSSIAN KERNEL DENSITY ESTIMATOR
スポンサーリンク
概要
- 論文の詳細を見る
It seems that the curve of the Gaussian kernel density estimate is often flat as compared with that of true p.d.f., although its window width has been chosen suitably. This flatness sometimes causes a bad kernel estimate of density. In this paper we shall introduce a scale parameter into true p.d.f. to show the flatness of the curve of Gaussian kernel density estimate. We study the asymptotic properties of the scale parameter to propose a modified kernel density estimator. Some simulations show a superiority of the estimator.
- Research Association of Statistical Sciencesの論文
- 1996-03-00
Research Association of Statistical Sciences | 論文
- CLUSTERING BY A FUZZY METRIC : APPLICATIONS TO THE CLUSTER-MEDIAN PROBLEM
- A FAMILY OF REGRESSION MODELS HAVING PARTIALLY ADDITIVE AND MULTIPLICATIVE COVARIATE STRUCTURE
- AN OPTIMAL STOPPING PROBLEM ON TREE
- ON THE ORDERS OF MAX-MIN FUNCTIONALS
- TREE EXPRESSIONS AND THEIR PRODUCT FORMULA