×

Testing for lack of fit in inverse regression-with applications to biophotonic imaging. (English) Zbl 1231.62060

Summary: We propose two test statistics for use in inverse regression problems \(Y=K\theta +\varepsilon \), where \(K\) is a given linear operator which cannot be continuously inverted. Thus, only noisy, indirect observations \(Y\) for the function \(\theta \) are available. Both test statistics have a counterpart in classical hypothesis testing, where they are called the order selection test and the data-driven Neyman smooth test. We also introduce two model selection criteria which extend the classical Akaike information criterion and Bayes information criterion to inverse regression problems. In a simulation study we show that the inverse order selection and Neyman smooth tests outperform their direct counterparts in many cases. The theory is motivated by data arising in confocal fluorescence microscopy. Here, images are observed with blurring, modelled as convolution, and stochastic error at subsequent times. The aim is then to reduce the signal-to-noise ratio by averaging over the distinct images. In this context it is relevant to decide whether the images are still equal, or have changed by outside influences such as moving of the object table.

MSC:

62G08 Nonparametric regression and quantile regression
62G10 Nonparametric hypothesis testing
62P10 Applications of statistics to biology and medical sciences; meta analysis
65C60 Computational problems in statistics (MSC2010)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Aerts, Testing the fit of a parametric function, J. Am. Statist. Ass. 94 pp 869– (1999) · Zbl 0996.62044
[2] Aerts, Testing lack of fit in multiple regression, Biometrika 87 pp 405– (2000) · Zbl 0948.62028
[3] Akaike, Proc. 2nd Int. Symp. Information Theory pp 267– (1973)
[4] Bewersdorf, Handbook of Biological Confocal Microscopy pp 561– (2006)
[5] Bhattacharya, Normal Approximation and Asymptotic Expansions (1976)
[6] Billingsley, Convergence of Probability Measures (1968)
[7] Bissantz, Convergence rates of general regularization methods for statistical inverse problems, SIAM J. Numer. Anal. 45 pp 2610– (2007) · Zbl 1234.62062
[8] Bissantz, Parametric versus non-parametric modeling?: statistical evidence based on P-value curves, Mnthly Notes R. Astronom. Soc. 340 pp 1190– (2003)
[9] Butucea, Adaptive estimation of linear functionals in the convolution model and applications (2007)
[10] Cavalier, Efficient estimation of a density in a problem of tomography, Ann. Statist. 28 pp 630– (2000) · Zbl 1105.62331
[11] Cavalier, Adaptive estimation for inverse problems with noisy operators, Inv. Probl. 21 pp 1345– (2005) · Zbl 1074.62053
[12] Cavalier, Sharp adaptation for inverse problems with random noise, Probab. Theory Reltd Flds 123 pp 323– (2002)
[13] Chow, Probability Theory: Independence, Interchangeability, Martingales (1978)
[14] Claeskens, Goodness of fit via nonparametric likelihood ratios, Scand. J. Statist. 31 pp 487– (2004) · Zbl 1065.62056
[15] Eubank, Testing goodness-of-fit in regression via order selection criteria, Ann. Statist. 20 pp 1412– (1992) · Zbl 0776.62045
[16] Fan, On the optimal rates of convergence for nonparametric deconvolution problems, Ann. Statist. 19 pp 1257– (1991) · Zbl 0729.62033
[17] Goldenshluger, Recovering convex edges of an image from noisy tomographic data, IEEE Trans. Inform. Theory 52 pp 1322– (2006) · Zbl 1309.94030
[18] Hart, Nonparametric Smoothing and Lack-of-fit Tests (1997) · Zbl 0886.62043
[19] Hoffmann, Nonlinear estimation for linear inverse problems with error in the operator, Ann. Statist. 36 pp 310– (2007)
[20] Holzmann, Density testing in a contaminated sample, J. Multiv. Anal. 98 pp 57– (2007) · Zbl 1102.62045
[21] Inglot, Intermediate approach to comparison of some goodness-of-fit tests, Ann. Inst. Statist. Math. 53 pp 810– (2001) · Zbl 1003.62042
[22] Inglot, Data-driven score tests for homoscedastic linear regression model: asymptotic results, Probab. Math. Statist. 26 pp 41– (2006) · Zbl 1104.62051
[23] Johnstone, Wavelet deconvolution in a periodic setting, J. R. Statist. Soc. B 66 pp 547– (2004) · Zbl 1046.62039
[24] Johnstone, Speed of estimation in positron emission tomography and related inverse problems, Ann. Statist. 18 pp 251– (1990) · Zbl 0699.62043
[25] Kaipio, Computational and Statistical Methods for Inverse Problems (2005)
[26] Kallenberg, Consistency and Monte Carlo simulation of a data driven version of smooth goodness-of-fit tests, Ann. Statist. 23 pp 1594– (1995) · Zbl 0847.62035
[27] Kress, Linear Integral Equations (1999)
[28] Ledwina, Data-driven version of Neyman’s smooth test of fit, J. Am. Statist. Ass. 89 pp 1000– (1994) · Zbl 0805.62022
[29] Loubes, Penalized estimators for nonlinear inverse problems (2004)
[30] Mair, Statistical inverse estimation in Hilbert scales, SIAM J. Appl. Math. 56 pp 1424– (1996) · Zbl 0864.62020
[31] Mallows, Some comments on Cp, Technometrics 15 pp 661– (1973) · Zbl 0269.62061
[32] Munk, On difference-based variance estimation in nonparametric regression when the covariate is high dimensional, J. R. Statist. Soc. B 67 pp 19– (2005) · Zbl 1060.62047
[33] Natterer, The Mathematics of Computerized Tomography (1986)
[34] Neyman, ’Smooth’ test for goodness of fit, Skand. Akt. 20 pp 149– (1937) · JFM 63.1092.02
[35] Nourrit, Blind deconvolution for high-resolution confocal scanning laser ophthalmoscopy, J. Opt. A 7 pp 585– (2005)
[36] Pepperkok, High-throughput fluorescence microscopy for systems biology, Nat. Rev. Molec. Cell. Biol. 9 pp 690– (2006)
[37] Petrov, Sequences of Independent Random Variables (1995)
[38] Rice, Bandwidth choice for nonparametric regression, Ann. Statist. 12 pp 1215– (1984) · Zbl 0554.62035
[39] Schwarz, Estimating the dimension of a model, Ann. Statist. 6 pp 461– (1978) · Zbl 0379.62005
[40] Spitzer, A combinatorial lemma and its applications to probability theory, Trans. Am. Math. Soc. 82 pp 323– (1956) · Zbl 0071.13003
[41] Vardi, A statistical model for positron emission tomography, J. Am. Statist. Ass. 80 pp 8– (1985) · Zbl 0561.62094
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.