Nonparametric estimates of low bias

zbMATH Open1297.62078arXiv1008.0127MaRDI QIDQ2921607FDOQ2921607

Authors: Christopher S. Withers, Saralees Nadarajah 

Publication date: 13 October 2014

Published in: REVSTAT (Search for Journal in Brave)

Abstract: We consider the problem of estimating an arbitrary smooth functional of $kgeq1$ distribution functions (d.f.s.) in terms of random samples from them. The natural estimate replaces the d.f.s by their empirical d.f.s. Its bias is generally $sim{n}^{-1}$, where $n$ is the minimum sample size, with a {it $p$th order} iterative estimate of bias $sim{n}^{-p}$ for any $p$. For $pleq4$, we give an explicit estimate in terms of the first $2p-2$ von Mises derivatives of the functional evaluated at the empirical d.f.s. These may be used to obtain {it unbiased} estimates, where these exist and are of known form in terms of the sample sizes; our form for such unbiased estimates is much simpler than that obtained using polykays and tables of the symmetric functions. Examples include functions of a mean vector (such as the ratio of two means and the inverse of a mean), standard deviation, correlation, return times and exceedances. These $p$th order estimates require only $simn$ calculations. This is in sharp contrast with computationally intensive bias reduction methods such as the $p$th order bootstrap and jackknife, which require $sim{n}^p$ calculations.

Full work available at URL: https://arxiv.org/abs/1008.0127

Recommendations

• Bias reduction by taylor series^∗
• scientific article; zbMATH DE number 4215162
• Estimates of low bias for the multivariate normal
• Delta and jackknife estimators with low bias for functions of binomial and multinomial parameters
• Density estimates of low bias

zbMATH Keywords

correlationmultivariatenonparametricbias reductionstandard deviationexceedancesreturn timesratio of meansvon Mises derivativesmultisample

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Order statistics; empirical distribution functions (62G30)

Cited In (9)

• Delta and jackknife estimators with low bias for functions of binomial and multinomial parameters
• A note on bias reduction
• On multivariate higher order von Mises expansions
• Bias-reduced estimates for skewness, kurtosis, \(L\)-skewness and \(L\)-kurtosis
• A unified method for constructing expectation tolerance intervals
• Bias reduction when data are rounded
• Bias reduction by taylor series^∗
• Estimates of low bias for the multivariate normal
• On the Effect of Bias Estimation on Coverage Accuracy in Nonparametric Inference

This page was built for publication: Nonparametric estimates of low bias

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2921607)