Bandwidth choice for nonparametric classification
نویسندگان
چکیده
منابع مشابه
Bandwidth Choice for Nonparametric Classification
It is shown that, for kernel-based classification with univariate distributions and two populations, optimal bandwidth choice has a dichotomous character. If the two densities cross at just one point, where their curvatures have the same signs, then minimum Bayes risk is achieved using bandwidths which are an order of magnitude larger than those which minimize pointwise estimation error. On the...
متن کاملBandwidth choice for robust nonparametric scale function estimation
Some key words: Cross–validation; Data–driven bandwidth; Heteroscedasticity; Local M−estimators; Nonparametric regression; Robust estimation. Corresponding Author Graciela Boente Instituto de Cálculo Facultad de Ciencias Exactas y Naturales Ciudad Universitaria, Pabellón 2 Buenos Aires, C1428EHA Argentina email: [email protected] fax 54-11-45763375 Running Head: Robust scale estimation. ∗This r...
متن کاملVariable data driven bandwidth choice in nonparametric quantile regression
The choice of a smoothing parameter or bandwidth is crucial when applying nonparametric regression estimators. In nonparametric mean regression various methods for bandwidth selection exists. But in nonparametric quantile regression bandwidth choice is still an unsolved problem. In this paper a selection procedure for local varying bandwidths based on the asymptotic mean squared error (MSE) of ...
متن کاملPenalizing function based bandwidth choice in nonparametric quantile regression
Abstract: In nonparametric mean regression various methods for bandwidth choice exist. These methods can roughly be divided into plug-in methods and methods based on penalizing functions. This paper uses the approach based on penalizing functions and adapt it to nonparametric quantile regression estimation, where bandwidth choice is still an unsolved problem. Various criteria for bandwitdth cho...
متن کاملBandwidth selection for nonparametric kernel testing
We propose a sound approach to bandwidth selection in nonparametric kernel testing. The main idea is to find an Edgeworth expansion of the asymptotic distribution of the test concerned. Due to the involvement of a kernel bandwidth in the leading term of the Edgeworth expansion, we are able to establish closed–form expressions to explicitly represent the leading terms of both the size and power ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The Annals of Statistics
سال: 2005
ISSN: 0090-5364
DOI: 10.1214/009053604000000959