Abstract
In this article we consider a sequence of hierarchical space model of inverse problems. The underlying function is estimated from indirect observations over a variety of error distributions including those that are heavy-tailed and may not even possess variances or means. The main contribution of this paper is that we establish some oracle inequalities for the inverse problems by using quantile coupling technique that gives a tight bound for the quantile coupling between an arbitrary sample p-quantile and a normal variable, and an automatic selection principle for the nonrandom filters. This leads to the data-driven choice of weights. We also give an algorithm for its implementation. The quantile coupling inequality developed in this paper is of independent interest, because it includes the median coupling inequality in literature as a special case.
Similar content being viewed by others
References
Bártfai P. Über die Entfernung der Irrfahrtswege. Studia Sci Math Hungar, 1970, 5: 41–49
Brown L D, Cai T T, Zhou H H. Robust nonparametric estimation via wavelet median regression. Ann Statist, 2008, 36: 2055–2084
Burnashev M V. Asymptotic expansions for median estimate of a parameter. Theory Probab Appl, 1996, 41: 632–645
Cai T. Adaptive wavelet estimation: A block thresholding and oracle inequality approach. Ann Statist, 1999, 27: 898–924
Candès E J, Donoho D L. Recovering edges in ill-posed inverse problems: Optimality of curvelet frames. Ann Statist, 2002, 30: 784–842
Cavalier L, Golubev G K, Picard D, et al. Oracle inequalities for inverse problems. Ann Statist, 2002, 30: 843–874
Cavalier L, Tsybakov A. Sharp adaptation for inverse problems with random noise. Probab Theory Related Fields, 2002, 123: 323–354
Csörgő M, Révész P. A new method to prove Strassen type laws of invariance principle I–II. ZWahrscheinlichkeitstheorie verw Geb, 1975, 31: 255–259 and 261–269
Cramér H. Mathematical Methods of Statistics. Princeton, NJ: Princeton University Press, 1946
Johnstone I M. Wavelet shrinkage for correlated data and inverse problems: Adaptivity results. Statist Sinica, 1999, 9: 51–83
Kolmogorov A N. Method of median in the theory of errors. Mat Sb, 1931, 38: 47–50
Komlós J, Major P, Tusnády G. An approximation of partial sums of independent RV’-s and the sample DF. I. Z Wahrscheinlichkeitstheorie verw Geb, 1975, 32: 111–131
Komlós J, Major P, Tusnády G. An approximation of partial sums of independent RV’-s and the sample DF. II. Z Wahrscheinlichkeitstheorie verw Geb, 1976, 34: 33–58
Hall P, Patil P. On the choice of smoothing parameter, threshold and truncation in nonparametric regression by wavelet methods. J Roy Statist Soc Ser B, 1996, 58: 361–377
Hájek J, Šidák Z. Theory of Rank Tests. Prague: CSAV, 1967
Pinsker M S. Optimal filtering of square integrable signals in Gaussian white noise. Problems Inform Transmission, 1980, 16: 120–133
Pollard D P. A User’s Guide to Measure Theoretic Probability. Cambridge: Cambridge University Press, 2001
Reis R D. Approximate Distributions of Order Statistics with Applications to Nonparametric Statistics. New York: Springer-Verlag, 1989
Stuck B W, Kleiner B. A statistical analysis of telephone noise. Bell System Technical J, 1974, 53: 1263–1320
Stuck B W. A historical overview of stable probability distributions in signal processing. In: IEEE International Conference on Acoustics, Speech, and Signal Processing. Washington: IEEE, 2000, 3795–3797
Tian M Z, Chan N H. Saddle point approximation and volatility estimation of value-at-risk. Statist Sinica, 2010, 20: 1239–1256
Tian M Z, Chen G M. Hierarchical linear regression models for conditional quantiles. Sci China Ser A, 2006, 49: 11–16
Tian M Z, Tang M L, Chan P S. Semiparametric quantile modelling of hierarchical data. Acta Math Sin, 2009, 25: 597–616
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tian, M. Robust estimation in inverse problems via quantile coupling. Sci. China Math. 55, 1029–1041 (2012). https://doi.org/10.1007/s11425-012-4365-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-012-4365-3
Keywords
- inverse problem
- robust estimation
- oracle inequalities
- quantile coupling inequalities
- heavy-tailed distributions
- hierarchical sequence space model