Journal of Statistical Theory and Practice

, Volume 11, Issue 3, pp 375–392 | Cite as

On the accuracy of fixed sample and fixed width confidence intervals based on the vertically weighted average

  • Ansgar StelandEmail author


Vertically weighted averages perform a bilateral filtering of data, in order to preserve fine details of the underlying signal, especially discontinuities such as jumps (in one dimension) or edges (in two dimensions). In homogeneous regions of the domain the procedure smoothes the data by averaging nearby data points to reduce the noise, whereas in inhomogeneous regions the neighboring points are only taken into account when their value is close to the current one. This results in a denoised reconstruction or estimate of the true signal without blurring finer details. This article addresses the lack of results about the construction and evaluation of confidence intervals based on the vertically weighted average, which is required for a proper statistical evaluation of its estimation accuracy. Based on recent results we discuss and investigate in greater detail fixed sample and fixed width (conditional) confidence intervals constructed from this estimator. The fixed width approach allows to specify explicitly the estimator’s accuracy and determines a random sample size to ensure the required coverage probability. This also fixes to some extent the inherent property of the vertically weighted average that its variability is higher in low-density regions than in high-density regions. To estimate the variances required to construct the procedures, we rely on resampling techniques, especially the bootstrap and the jackknife. Extensive Monte Carlo simulations show that, in general, the proposed confidence intervals are highly reliable in terms of their coverage probabilities for a wide range of parameter settings. The performance can be further increased by the bootstrap.


Bilateral filter bootstrap jackknife jump-preserving estimation two stage confidence interval signal estimation 

AMS Subject Classification

62L10 62G09 62G15 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Donoho, D. L., and I. M. Johnstone. 1994. Ideal spatial adaptation by wavelet shrinkage. Biometrika 81(3):425–55,MathSciNetCrossRefGoogle Scholar
  2. Friedrich F., A. Kempe, V. Liebscher, and G. Winkler. 2008. Complexity penalized M -estimation: Fast computation. Journal of Computational and Graphical Statistics 17 (1):201–24.MathSciNetCrossRefGoogle Scholar
  3. Mukhopadhyay, N. 1980. A consistent and asymptotically efficient two-stage procedure to construct fixed width confidence intervals for the mean. Metrika 27(4):281–84.MathSciNetCrossRefGoogle Scholar
  4. Mukhopadhyay, N., and B. M. de Silva. 2009. Sequential Methods and Their Applications. Boca Raton, FL: Taylor & Francis.zbMATHGoogle Scholar
  5. Pawlak, M., and E. Rafajłowicz. 2000. Vertically weighted regression — A tool for non-linear data analysis and constructing control charts. Journal of the German Statistical Society 84(4):367–88.MathSciNetzbMATHGoogle Scholar
  6. Pawlak, M., E. Rafajłowicz, and A. Steland. 2004. On detecting jumps in time series: Nonparametric setting. Journal of Nonparametric Statistics 16(3–4):329–47.MathSciNetCrossRefGoogle Scholar
  7. Pawlak, M., and A. Steland. 2013. Nonparametric sequential signal change detection under dependent noise. IEEE Trans. Information Theory 59 (6):3514–31.MathSciNetCrossRefGoogle Scholar
  8. Polzehl, J., and V. Spokoiny. 2003. Image denoising: Pointwise adaptive approach. Annals of Statistics 31 (1):30–57.MathSciNetCrossRefGoogle Scholar
  9. Qiu, P. 1998. Discontinuous regression surfaces fitting. Annals of Statistics 26 (6):2218–45.MathSciNetCrossRefGoogle Scholar
  10. Qiu, P. 2005. Image Processing and Jump Regression Analysis. Wiley Series in Probability and Statistics Hoboken, NJ: Wiley-Interscience.CrossRefGoogle Scholar
  11. Quenouille, M. 1949. Approximate tests of correlation in time series. Journal of the Royal Statistical Society — Series B 11:68–84.MathSciNetzbMATHGoogle Scholar
  12. Rafajłowicz, E. 2007. SUSAN edge detector reinterpreted, simplified and modified. In 2007 International Workshop on Multidimensional (nD) Systems, 69–74, June.Google Scholar
  13. Rafajłowicz, E., M. Pawlak, and A. Steland. 2008. Nonlinear image processing and filtering: a unified approach based on vertically weighted regression. International Journal of Applied Mathematics and Computer Science 18 (1):49–61.MathSciNetCrossRefGoogle Scholar
  14. Rafajłowicz, E., M. Pawlak, and A. Steland. 2010. Nonparametric sequential change-point detection by a vertically trimmed box method. IEEE Trans. Information Theory 56 (7):3621–34.MathSciNetCrossRefGoogle Scholar
  15. Rafajłowicz, E., and A. Steland. 2009. A binary control chart to detect small jumps. Statistics 43 (3):295–311.MathSciNetCrossRefGoogle Scholar
  16. Steland, A. 2005. On the distribution of the clipping median under a mixture model. Statistics Probability Letters 71 (1):1–13.MathSciNetCrossRefGoogle Scholar
  17. Steland, A. 2015. Vertically weighted averages in Hilbert spaces and applications to imaging: fixed-sample asymptotics and efficient sequential two-stage estimation. Sequential Analysis 34 (3):295–323.MathSciNetCrossRefGoogle Scholar
  18. Tomasi, C., and R. Manduchi. 1998. Bilateral filtering for gray and color images. In Proceedings of the 1998 IEEE International Conference on Computer Vision, 839–46, June.Google Scholar
  19. Tukey, J. W. 1958. Bias and confidence in not quite large samples [Abstract]. Annals of Mathematical Statistics 60 (29):614.Google Scholar
  20. Wand, M. P., and M. C. Jones. 1995. Kernel Smoothing, vol. 60 of Monographs on Statistics and Applied Probability. London, UK: Chapman and Hall, Ltd.Google Scholar

Copyright information

© Grace Scientific Publishing 2017

Authors and Affiliations

  1. 1.Institute of StatisticsRWTH Aachen UniversityAachenGermany

Personalised recommendations