A geometric approach to density estimation with additive noise.We introduce and study a method for density estimation under an additive noise model. Our method does not attempt to maximize a likelihood, but rather is purely geometric: heuristically, we L 2 -project the observed empirical distribution onto the space of candidate densities that are reachable under the additive noise model. Our estimator reduces to a quadratic program, and so can be computed efficiently. In simulation studies, it roughly matches the accuracy of fully general maximum likelihood estimators at a fraction of the computational cost. We give a theoretical analysis of the estimator and show that it is consistent, attains a quasi-parametric convergence rate under moment conditions, and is robust to model mis-specification. We provide an R implementation of the proposed estimator in the package nlpden.
Keywords for this software
References in zbMATH (referenced in 3 articles )
Showing results 1 to 3 of 3.
- Madrid-Padilla, Oscar-Hernan; Polson, Nicholas G.; Scott, James: A deconvolution path for mixtures (2018)
- Pan, Jia-Chiun; Huang, Yufen; Hwang, J. T. Gene: Estimation of selected parameters (2017)
- Wager, Stefan: A geometric approach to density estimation with additive noise (2014)