Abstract
The classic superefficient estimate of Hodges for a one dimensional normal mean and the modern hard thresholding estimates introduced in the works of David Donoho and Iain Johnstone exhibit some well known risk phenomena. They provide quantifiable improvement over the MLE near zero, but also suffer from risk inflation suitably away from zero. Classic work of Le Cam and Hájek has precisely pinned down certain deep and fundamental aspects of these risk phenomena.
In this article, we study risks and Bayes risks of general thresholding estimates. In particular, we show that reversal to the risk of the MLE occurs at one standard deviation from zero, but the global peak occurs in a small lower neighborhood of the thresholding parameter. We give first order limits and various higher order asymptotic expansions to quantify these phenomena. Separately, we identify those priors in a class under which the thresholding estimate would be Bayesianly preferred to the MLE and use the theory of regular variation to pin down the rate at which the difference of their Bayes risks goes to zero.
We also formulate and answer an optimal thresholding question, which asks for the thresholding estimate that minimizes the global maximum of the risk subject to a specified gain at zero.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Banerjee M (2006) Superefficiency, contiguity, LAN University Michigan Lecture Notes
Bickel PJ (1983) Minimax estimation of the mean of a normal distribution subject to doing well at a point. In: Recent Advances in Statistics, A Festschrift for H. Chernoff, M. Rizvi, J. Rustagi and D. Siegmund eds., 511-528, Academic Press, NY
DasGupta A (2008) Asymptotic theory of statistics and probability. Springer, New York
DasGupta A (2011) Probability for statistics and machine learning: fundamentals and advanced topics. Springer, New York
Donoho D, Johnstone IM (1994) Ideal spatial adaptation by wavelet shrinkage. Biom 81:425–455
Hájek J (1970) A characterization of limiting distributions of regular estimates, Z. Wahr verw Geb 14: 323-330
Jeganathan P (1983) Some asymptotic properties of risk functions when the limit of the experiment is mixed normal. Sankhy\(\bar a\) Ser A 45:66–87
Johnstone IM (2012) Function estimation and Gaussian sequence models, Cambridge University Press, Forthcomiing
Le Cam L (1953) On some asymptotic properties of maximum likelihood estimates and related Bayes estimates. University California Publications 1: 277-330
Le Cam L (1973) Sur les contraintes imposees par les passages a la limite usuels en statistique, Proc. 39th session International Statistical Institute, XLV, 169-177
Lehmann EL, Romano J (2005) Testing statistical hypotheses, 3rd edn. Springer, New York
van der Vaart A (1997) Superefficiency, in Festschrift for Lucien Le Cam, D. Pollard, E. Torgersen, G. Yang, Eds. Springer, New York
van der Vaart A (1998) Asymptotic statistics. Cambridge University Press, Cambridge
Wasserman L (2005) All of Nonparametric Statistics. Springer, New York
Wellner J (2012) University Washington Lecture Notes
Acknowledgement
It is a pleasure to thank Jyotishka Datta for his very gracious help with formatting the paper.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
DasGupta, A., Johnstone, I. (2014). Asymptotic Risk and Bayes Risk of Thresholding and Superefficient Estimates and Optimal Thresholding. In: Lahiri, S., Schick, A., SenGupta, A., Sriram, T. (eds) Contemporary Developments in Statistical Theory. Springer Proceedings in Mathematics & Statistics, vol 68. Springer, Cham. https://doi.org/10.1007/978-3-319-02651-0_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-02651-0_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-02650-3
Online ISBN: 978-3-319-02651-0
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)