Lower bounds for bandwidth selection in density estimation (Q1178978)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Lower bounds for bandwidth selection in density estimation
scientific article

    Statements

    Lower bounds for bandwidth selection in density estimation (English)
    0 references
    0 references
    26 June 1992
    0 references
    It is shown that any bandwidth selector for a kernel density estimator is at least \(n^{-1/10}\) apart (in relative magnitude) from the minimizer of the integrated squared error, while for the mean integrated squared error the lower bound is \(n^{-1/2}\). This bound is achieved for certain plug-in estimators.
    0 references
    0 references
    relative rate of convergence
    0 references
    cross-validation
    0 references
    bandwidth selector
    0 references
    kernel density estimator
    0 references
    integrated squared error
    0 references
    mean integrated squared error
    0 references
    lower bound
    0 references
    plug-in estimators
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references