Answers are grounded in this paper and cite their sources.
A new approach to <tex-math notation="LaTeX">L₂ </tex-math> -consistent estimation of a general density functional using <tex-math notation="LaTeX">k </tex-math> -nearest neighbor distances is proposed, where the functional under consideration is in the form of the expectation of some function <tex-math notation="LaTeX">f </tex-math> of the densities at each point. The estimator is designed to be asymptotically unbiased, using the convergence of the normalized volume of a <tex-math notation="LaTeX">k </tex-math> -nearest neighbor ball to a Gamma distribution in the large-sample limit, and naturally involves the inverse Laplace transform of a scaled version of the function <tex-math notation="LaTeX">f </tex-math> . Some instantiations of the proposed estimator recover existing <tex-math notation="LaTeX">k </tex-math> -nearest neighbor based estimators of Shannon and Rényi entropies and Kullback–Leibler and Rényi divergences, and discover new consistent estimators for many other functionals such as logarithmic entropies and divergences. The <tex-math notation="LaTeX">L₂ </tex-math> -consistency of the proposed estimator is established for a broad class of densities for general functionals, and the convergence rate in mean squared error is established as a function of the sample size for smooth, bounded densities.
No takes yet. Share an insight, caveat, or question.
Ryu et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: