Multichannel Deconvolution with Long-Range Dependence: A Minimax Study

We consider the problem of estimating the unknown response function in the multichannel deconvolution model with long-range dependent Gaussian errors. We do not limit our consideration to a specific type of long-range dependence rather we assume that the errors should satisfy a general assumption in terms of the smallest and larger eigenvalues of their covariance matrices. We derive minimax lower bounds for the quadratic risk in the proposed multichannel deconvolution model when the response function is assumed to belong to a Besov ball and the blurring function is assumed to possess some smoothness properties, including both regular-smooth and super-smooth convolutions. Furthermore, we propose an adaptive wavelet estimator of the response function that is asymptotically optimal (in the minimax sense), or near-optimal within a logarithmic factor, in a wide range of Besov balls. It is shown that the optimal convergence rates depend on the balance between the smoothness parameter of the response function, the kernel parameters of the blurring function, the long memory parameters of the errors, and how the total number of observations is distributed among the total number of channels. Some examples of inverse problems in mathematical physics where one needs to recover initial or boundary conditions on the basis of observations from a noisy solution of a partial differential equation are used to illustrate the application of the theory we developed. The optimal convergence rates and the adaptive estimators we consider extend the ones studied by Pensky and Sapatinas (2009, 2010) for independent and identically distributed Gaussian errors to the case of long-range dependent Gaussian errors.

[1]  G. Walter,et al.  Deconvolution Using Meyer Wavelets , 1999 .

[2]  Roman Vershynin,et al.  Introduction to the non-asymptotic analysis of random matrices , 2010, Compressed Sensing.

[3]  Theofanis Sapatinas,et al.  Functional deconvolution in a periodic setting: Uniform case , 2007 .

[4]  Peter McCullagh.  [The Unifying Role of Iterative Generalized Least Squares in Statistical Algorithms]: Comment , 1989 .

[5]  Jan Beran,et al.  Statistics for long-memory processes , 1994 .

[6]  Simon J. Godsill,et al.  Dicussion on the meeting on ‘Statistical approaches to inverse problems’ , 2004 .

[7]  Gerard Kerkyacharian,et al.  Wavelet deconvolution in a periodic setting , 2004 .

[8]  B. Vidakovic,et al.  Adaptive wavelet estimator for nonparametric density deconvolution , 1999 .

[9]  THE SMOOTHING DICHOTOMY IN RANDOM-DESIGN REGRESSION WITH LONG-MEMORY ERRORS BASED ON MOVING AVERAGES , 2000 .

[10]  R. Khas'minskii,et al.  A STATISTICAL APPROACH TO SOME INVERSE PROBLEMS FOR PARTIAL DIFFERENTIAL EQUATIONS , 1999 .

[11]  Minimax estimation via wavelets for indirect long-memory data , 1997 .

[12]  Murad S. Taqqu,et al.  Theory and applications of long-range dependence , 2003 .

[13]  T. Sapatinas,et al.  Multichannel boxcar deconvolution with growing number of channels , 2011, 1102.2298.

[14]  Robert Lattès,et al.  Méthode de quasi-réversibilbilité et applications , 1967 .

[15]  Dominique Picard,et al.  ADAPTIVE BOXCAR DECONVOLUTION ON FULL LEBESGUE MEASURE SETS , 2007 .

[16]  Jan Beran,et al.  Long-Memory Processes: Probabilistic Properties and Statistical Methods , 2013 .

[17]  Wolfgang Heardle et al..  Wavelets, approximation, and statistical applications , 2013 .

[18]  A. Tsybakov,et al.  Aggregation for Gaussian regression , 2007, 0710.3654.

[19]  D. Donoho.  Nonlinear Solution of Linear Inverse Problems by Wavelet–Vaguelette Decomposition , 1995 .

[20]  Rafal Kulik.  Nonparametric deconvolution problem for dependent sequences , 2007, 0711.4004.

[21]  S W Dho,et al.  Deconvolution of long-pulse lidar signals with matrix formulation. , 1997, Applied optics.

[22]  U. Grenander,et al.  Toeplitz Forms And Their Applications , 1958 .

[23]  P. Robinson,et al.  Semiparametric estimation from time series with long-range dependence , 1994 .

[24]  Alexandre B. Tsybakov,et al.  Introduction to Nonparametric Estimation , 2008, Springer series in statistics.

[25]  Yu. Golubev.  The principle of penalized empirical risk in severely ill-posed problems , 2004 .

[26]  Simultaneous Wavelet Deconvolution in Periodic Setting , 2006 .

[27]  Rainer Reuter,et al.  STABLE DECONVOLUTION OF NOISY LIDAR SIGNALS , 2000 .

[28]  L p -WAVELET REGRESSION WITH CORRELATED ERRORS AND INVERSE PROBLEMS , 2008 .

[29]  Block thresholding for wavelet-based estimation of function derivatives from a heteroscedastic multichannel convolution model , 2012, 1202.6316.

[30]  H. Zanten,et al.  A note on wavelet density deconvolution for weakly dependent data , 2008 .

[31]  J. Mielniczuk,et al.  ON RANDOM-DESIGN MODEL WITH DEPENDENT ERRORS , 2004 .

[32]  J. Wishart.  Wavelet deconvolution in a periodic setting with long-range dependent errors , 2012, 1208.4441.

[33]  I. S. Gradshteyn,et al.  Table of Integrals, Series, and Products , 1976 .

[34]  Ja-Yong Koo,et al.  Wavelet deconvolution , 2002, IEEE Trans. Inf. Theory.

[35]  I. Johnstone,et al.  Periodic boxcar deconvolution and diophantine approximation , 2004, math/0503663.

[36]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[37]  F. Comte,et al.  Adaptive density deconvolution with dependent inputs , 2006, math/0606166.

[38]  B. Silverman,et al.  Wavelet decomposition approaches to statistical inverse problems , 1998 .

[39]  J. Geweke,et al.  THE ESTIMATION AND APPLICATION OF LONG MEMORY TIME SERIES MODELS , 1983 .

[40]  M. Rudelson,et al.  Hanson-Wright inequality and sub-gaussian concentration , 2013 .

[41]  A. Tsybakov,et al.  Wavelets, approximation, and statistical applications , 1998 .

[42]  D. F. Hays,et al.  Table of Integrals, Series, and Products , 1966 .

[43]  S. Mallat,et al.  Thresholding estimators for linear inverse problems and deconvolutions , 2003 .

[44]  T. Sapatinas,et al.  A ug 2 00 9 On Convergence Rates Equivalency and Sampling Strategies in a Functional Deconvolution Model , 2009 .

[45]  J. Beran.  Statistical methods for data with long-range dependence , 1992 .

[46]  T. Sapatinas,et al.  Functional Deconvolution in a Periodic Setting , 2007 .