On-line density estimators with high efficiency

Summary, in English

We present on-line procedures for estimating density functions and their derivatives. At each step, M terms are updated. By increasing M the efficiency compared to the traditional off-line kernel density estimator tends to one. Already for M = 2, it exceeds 99.1% for kernel orders and derivatives of practical interest.

• Författare: Ola Hössjer; Ulla Holst
• Avdelning/ar: Matematisk statistik
• Publiceringsår: 1995
• Språk: Engelska
• Sidor: 829-833
• Publikation/Tidskrift/Serie: IEEE Transactions on Information Theory
• Volym: 41
• Issue: 3
• Länkar: http://ieeexplore.ieee.org/iel1/18/8656/00382036.pdf?tp=&arnumber=382036&isnumber=8656; http://dx.doi.org/10.1109/18.382036
• Dokumenttyp: Artikel i tidskrift
• Förlag: IEEE - Institute of Electrical and Electronics Engineers Inc.
• Ämne: Probability Theory and Statistics
• Nyckelord: recursive density estimator; Asymptotic mean-squared error; efficiency; kernel density estimator; on-line density estimator; on-line bandwidth selection
• Status: Published
• ISBN/ISSN/Övrigt: ISSN: 0018-9448