The Annals of Statistics

Deconvolution with unknown error distribution

Jan Johannes

Full-text: Open access

Abstract

We consider the problem of estimating a density fX using a sample Y1, …, Yn from fY=fXfε, where fε is an unknown density. We assume that an additional sample ε1, …, εm from fε is observed. Estimators of fX and its derivatives are constructed by using nonparametric estimators of fY and fε and by applying a spectral cut-off in the Fourier domain. We derive the rate of convergence of the estimators in case of a known and unknown error density fε, where it is assumed that fX satisfies a polynomial, logarithmic or general source condition. It is shown that the proposed estimators are asymptotically optimal in a minimax sense in the models with known or unknown error density, if the density fX belongs to a Sobolev space $H_{\mathcal{p}}$ and fε is ordinary smooth or supersmooth.

Article information

Source
Ann. Statist., Volume 37, Number 5A (2009), 2301-2323.

Dates
First available in Project Euclid: 15 July 2009

Permanent link to this document
https://projecteuclid.org/euclid.aos/1247663756

Digital Object Identifier
doi:10.1214/08-AOS652

Mathematical Reviews number (MathSciNet)
MR2543693

Zentralblatt MATH identifier
1173.62018

Subjects
Primary: 62G07: Density estimation 62G07: Density estimation
Secondary: 62G05: Estimation 42A38: Fourier and Fourier-Stieltjes transforms and other transforms of Fourier type

Keywords
Deconvolution Fourier transform kernel estimation spectral cut off Sobolev space source condition optimal rate of convergence

Citation

Johannes, Jan. Deconvolution with unknown error distribution. Ann. Statist. 37 (2009), no. 5A, 2301--2323. doi:10.1214/08-AOS652. https://projecteuclid.org/euclid.aos/1247663756


Export citation

References

  • [1] Bigot, J. and Van Bellegem, S. (2006). Log-density deconvolution by wavelet thresholding. Technical report, Univ. catholique de Louvain.
  • [2] Butucea, C. and Matias, C. (2005). Minimax estimation of the noise level and of the deconvolution density in a semiparametric deconvolution model. Bernoulli 11 309–340.
  • [3] Carrasco, M. and Florens, J.-P. (2002). Spectral method for deconvolving a density. Working Paper No. 138, Univ. Toulouse I, IDEI.
  • [4] Carroll, R. J. and Hall, P. (1988). Optimal rates of convergence for deconvolving a density. J. Amer. Statist. Assoc. 83 1184–1186.
  • [5] Cavalier, L. and Hengartner, N. (2005). Adaptive estimation for inverse problems with noisy operators. Inverse Problems 21 1345–1361.
  • [6] Comte, F., Rozenholc, Y. and Taupin, M.-L. (2006). Penalized contrast estimator for density deconvolution. Canad. J. Statist. 34 431–452.
  • [7] Devroye, L. (1989). Consistent deconvolution in density estimation. Canad. J. Statist. 17 235–239.
  • [8] Diggle, P. J. and Hall, P. (1993). A Fourier approach to nonparametric deconvolution of a density estimate. J. Roy. Statist. Soc. Ser. B 55 523–531.
  • [9] Efromovich, S. (1997). Density estimation for the case of supersmooth measurement error. J. Amer. Statist. Assoc. 92 526–535.
  • [10] Engl, H. W., Hanke, M. and Neubauer, A. (2000). Regularization of Inverse Problems. Kluwer Academic, Dordrecht.
  • [11] Fan, J. (1991). On the optimal rates of convergence for nonparametric deconvolution problems. Ann. Statist. 19 1257–1272.
  • [12] Fan, J. (1992). Deconvolution with supersmooth distributions. Canad. J. Statist. 20 734–747.
  • [13] Fan, J. and Koo, J. Y. (2002). Wavelet deconvolution. IEEE Trans. Inform. Theory 48 734–747.
  • [14] Goldenshluger, A. (1999). On pointwise adaptive nonparametric deconvolution. Bernoulli 5 907–925.
  • [15] Goldenshluger, A. (2000). Density deconvolution in the circular structural model. J. Multivariate Anal. 81 360–375.
  • [16] Hall, P. and Qiu, P. (2005). Discrete-transform approach to deconvolution problems. Biometrika 92 135–148.
  • [17] Hall, P. and Yao, Q. (2003). Inference in components of variance models with low replication. Ann. Statist. 31 414–441.
  • [18] Havilio, M. (2006). Signal deconvolution based expression-detection and background adjustment for microarray data. J. Comput. Biol. 13 63–80.
  • [19] Horowitz, J. L. and Markatou, M. (1996). Semiparametric estimation of regression models for panel data. Rev. Econom. Stud. 63 145–168.
  • [20] Kawata, T. (1972). Fourier Analysis in Probability Theory. Academic Press, New York.
  • [21] Kim, P. T. and Koo, J. Y. (2002). Optimal spherical deconvolution. J. Multivariate Anal. 80 21–42.
  • [22] Koo, J. Y. and Park, B. U. (1996). B-spline deconvolution based on the EM algorithm. J. Statist. Comput. Simulation 54 275–288.
  • [23] Levitt, D. G. (2003). The use of a physiologically based pharmacokinetic model to evaluate deconvolution measurements of systemic absorption. In BMC Clinical Pharmacology 3. BioMed Central. Available at http://www.biomedcentral.com/1472-6904/3/1.
  • [24] Liu, M. C. and Taylor, R. L. (1989). A consistent nonparametric density estimator for the deconvolution problem. Canad. J. Statist. 17 427–438.
  • [25] Mair, B. A. (1994). Tikhonov regularization for finitely and infinitely smoothing operators. SIAM J. Math. Anal. 25 135–147.
  • [26] Mair, B. A. and Ruymgaart, F. H. (1996). Statistical inverse estimation in Hilbert scales. SIAM J. Math. Anal. 56 1424–1444.
  • [27] Meister, A. (2006). Density estimation with normal measurement error with unknown variance. Statist. Sinica 16 195–211.
  • [28] Mendelsohn, J. and Rice, J. (1982). Deconvolution of micro-fluorometric histograms with B-splines. J. Amer. Statist. Assoc. 77 748–753.
  • [29] Nair, M., Pereverzev, S. V. and Tautenhahn, U. (2005). Regularization in Hilbert scales under general smoothing conditions. Inverse Problems 21 1851–1869.
  • [30] Natterer, F. (1984). Error bounds for Tikhonov regularization in Hilbert scales. Applicable Anal. 18 29–37.
  • [31] Neumann, M. H. (1997). On the effect of estimating the error density in nonparametric deconvolution. J. Nonparametr. Statist. 7 307–330.
  • [32] Neumann, M. H. (2006). Deconvolution from panel data with unknown error distribution. J. Multivariate Anal. 98 1955–1968.
  • [33] Parzen, E. (1962). On estimation of a probability density function and mode. Ann. Math. Statist. 33 1065–1076.
  • [34] Pensky, M. and Vidakovic, B. (1999). Adaptive wavelet estimator for nonparametric density deconvolution. Ann. Statist. 27 2033–2053.
  • [35] Petrov, V. V. (1995). Limit Theorems of Probability Theory. Sequences of Independent Random Variables, 4th ed. Clarendon Press, Oxford.
  • [36] Stefanski, L. A. (1990). Rate of convergence of some estimators in a class of deconvolution problems. Statist. Probab. Lett. 9 229–235.
  • [37] Tautenhahn, U. (1996). Error estimates for regularization methods in Hilbert scales. SIAM J. Numer. Anal. 33 2120–2130.
  • [38] Tessier, E. (1995). Analysis and calibration of natural guide star adaptive optics data. In Adaptive Optical Systems and Applications (R. K. Tyson and R. Q. Fugate, eds.) 178–193. Proc. SPIE 2534. Available at arXiv:astro-ph/9601174v1.
  • [39] Vainikko, G. M. and Veretennikov, A. Y. (1986). Iteration Procedures in Ill-Posed Problems. Nauka, Moscow (in Russian).
  • [40] van der Vaart, A. W. (1998). Asymptotic Statistics. Cambridge Univ. Press.
  • [41] Zhang, C.-H. (1990). Fourier methods for estimating mixing densities and distributions. Ann. Statist. 18 806–831.