The support reduction algorithm for computing non-parametric function estimates in mixture models. (English) Zbl 1199.65017

A new iterative algorithm (called support reduction, SR-algorithm) is described for minimization of convex functionals from bounded discrete positive measures with finite support. During each iteration, the algorithm adds one new support point to the existing iterate. After that, as many support points of the measure as possible are deleted resulting in a sparse next iterate. This algorithm is applied to the least squares estimation of convex regression function, to maximum likelihood deconvolution and to the statistical analysis of quantum non-locality experiments. The algorithm is compared to the EM-algorithm and the interior point algorithm.


65C60 Computational problems in statistics (MSC2010)
62G08 Nonparametric regression and quantile regression
62J02 General nonlinear regression


Full Text: DOI arXiv Link


[1] Aspect, Experimental test of Bell’s inequalities using time-varying analyzers, Phys. Rev. Lett. 49 pp 1804– (1982)
[2] Birke, Estimating a convex function in nonparametric regression, Scand. J. Statist. 34 pp 384– (2007) · Zbl 1142.62019
[3] Böhning, Convergence of Simar’s algorithm for finding the maximum likelihood estimate of a compound Poisson process, Ann. Statist. 10 pp 1006– (1982) · Zbl 0523.62074
[4] Böhning, A vertex-exchange method in D-optimal design theory, Metrika 33 pp 337– (1986)
[5] Böhning, A review of reliable maximum likelihood algorithms for semiparametric mixture models, J. Statist. Plan. Inference 47 pp 5– (1995)
[6] Clauser, Proposed experiment tot test local hidden-variables theories, Phys. Rev. Lett. 23 pp 880– (1969)
[7] Dam, The statistical strength of nonlocality proofs, IEEE Trans. Inf. Theory 51 pp 2812– (2005) · Zbl 1286.81017
[8] Dempster, Maximum likelihood from incomplete data via the EM algorithm, J. Roy. Statist. Soc. Ser. B 39 pp 1– (1977) · Zbl 0364.62022
[9] Dümbgen, L. , Hüsler, A. & Rufibach, K. (2007). Active set and EM algorithms for logconcave densities based on complete and censored data. arXiv:0707.4643.
[10] Gill, Asymptotics: Particles, Processes and Inverse Problems. Festschrift for Piet Groeneboom 55 pp 135– (2007)
[11] Groeneboom, A canonical process for estimation of convex functions: the ’invelope’ of integrated Brownian motion +t4, Ann. Statist. 29 pp 1620– (2001) · Zbl 1043.62026
[12] Groeneboom, Estimation of convex functions: characterizations and asymptotic theory, Ann. Statist. 29 pp 1653– (2001) · Zbl 1043.62027
[13] Groeneboom, The support reduction algorithm for computing nonparametric function estimates in mixture models (2003)
[14] Groeneboom, Current status data with competing risks: consistency and rates of convergence of the MLE, Ann. Statist. 36 (2007)
[15] Groeneboom, Information Bounds and Nonparametric Maximum Likelihood Estimation (1992) · Zbl 0757.62017
[16] Jongbloed, The iterative convex minorant algorithm for nonparametric estimation, J. Comp. Graph. Statist. 7 pp 310– (1998)
[17] Jongbloed, Estimating a concave distribution function from data corrupted with additive noise (2008)
[18] Jongbloed, Nonparametric inference for Lévy-driven Ornstein-Uhlenbeck processes, Bernoulli 11 pp 759– (2005) · Zbl 1084.62080
[19] Langaas, Estimating the proportion of true null hypotheses with application to DNA microarray data, J. Roy. Statist. Soc. Ser. B Statist. Methodol. 67 pp 555– (2005) · Zbl 1095.62037
[20] Lesperance, An algorithm for computing the nonparametric MLE of a mixing distribution, J. Amer. Statist. Assoc. 87 pp 120– (1992) · Zbl 0850.62336
[21] Lindsay, The geometry of mixture likelihoods: a general theory, Ann. Statist. 11 pp 86– (1983) · Zbl 0512.62005
[22] Luenberger, Introduction to Linear and Nonlinear Programming (1973)
[23] Maathuis, R package MLEcens (2007)
[24] Meyer, M. C. (1997). Shape restricted inference with applications to nonparametric regression, smooth nonparametric function estimation, and density estimation. Ph.D. dissertation. Department of Statistics, University of Michigan, Ann Arbor, Michigan.
[25] Robertson, Order Restricted Statistical Inference (1988)
[26] Simar, Maximum likelihood estimation of a compound Poisson process, Ann. Statist. 4 pp 1200– (1976) · Zbl 0362.62095
[27] Wright, Primal-Dual Interior-Point Methods (1994) · Zbl 0863.65031
[28] Zohren, On the maximal violation of the CGLMP inequality for infinite dimensional states (2006)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.