Abstract
The problem of estimating a probability density function f on the d−1-dimensional unit sphere S d−1 from directional data using the needlet frame is considered. It is shown that the decay of needlet coefficients supported near a point x∈S d−1 of a function f:S d−1→ℝ depends only on local Hölder continuity properties of f at x. This is then used to show that the thresholded needlet estimator introduced in Baldi, Kerkyacharian, Marinucci, and Picard (Ann. Stat. 39, 3362–3395, 2009) adapts to the local regularity properties of f. Moreover, an adaptive confidence interval for f based on the thresholded needlet estimator is proposed, which is asymptotically honest over suitable classes of locally Hölderian densities.
Similar content being viewed by others
References
Andersson: Characterization of pointwise Hölder regularity. Appl. Comput. Harmon. Anal. 4, 424–443 (1997)
Baldi, Kerkyacharian, Marinucci, Picard: Adaptive density estimation for directional data using needlets. Ann. Stat. 39, 3362–3395 (2009)
Donoho, Johnstone, Kerkyacharian, Picard: Density estimation by wavelet thresholding. Ann. Stat. 24, 508–539 (1996)
Faraut: Analysis on Lie Groups: An Introduction. Cambridge Studies in Advanced Mathematics, p. 189. Cambridge University Press, Cambridge (2008)
Geller, Mayeli: Continuous wavelets on compact manifolds. Math. Z. 262, 895–927 (2009)
Giné, Nickl: An exponential inequality for the distribution function of the kernel density estimator, with applications to adaptive estimation. Probab. Theory Relat. Fields 143, 569–596 (2009)
Giné, Nickl: Confidence bands in density estimation. Ann. Stat. 38, 1122–1170 (2010)
Hall, Watson, Cabrera: Kernel density estimation with spherical data. Biometrika 74, 751–762 (1987)
Hoffmann, Nickl: On adaptive inference and confidence bands. Ann. Stat. 39, 2383–2409 (2011)
Jaffard: Wavelet techniques for pointwise regularity. Ann. Fac. Sci. Toulouse XV, 3–33 (2006)
Kerkyacharian, Nickl, Picard: Concentration inequalities and confidence bands for needlet density estimators on compact homogeneous manifolds. Probab. Theory Relat. Fields. doi:10.1007/s00440-011-0348-5
Klemelä: Estimation of densities and derivatives of densities with directional data. J. Multivar. Anal. 73, 18–40 (2000)
Lepski, Mammen, Spokoiny: Optimal spatial adaptation to inhomogeneous smoothness: an approach based on kernel estimates with variable bandwidth selectors. Ann. Stat. 25, 929–947 (1997)
Low: On nonparametric confidence intervals. Ann. Stat. 25, 2547–2554 (1997)
Narcowich, Petrushev, Ward: Decomposition of Besov and Triebel-Lizorkin spaces on the sphere. J. Funct. Anal. 238, 530–564 (2006)
Narcowich, Petrushev, Ward: Localized tight frames on spheres. SIAM J. Math. Anal. 38, 574–594 (2006)
Picard, Tribouley: Adaptive confidence interval for pointwise curve estimation. Ann. Stat. 28, 298–335 (2000)
Stein: Weiss Introduction to Fourier Analysis on Euclidean Spaces, pp. 137–150. Princeton University Press, Princeton (1971)
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by G. Kerkyacharian.
This forms part of the author’s PHD thesis written under the supervision of Richard Nickl, whose expertise, understanding, and patience helped shape this paper. I also have much gratitude to the referees who wrote such extensive reports helping me to improve this paper. All mistakes remain mine.
Rights and permissions
About this article
Cite this article
Kueh, A. Locally Adaptive Density Estimation on the Unit Sphere Using Needlets. Constr Approx 36, 433–458 (2012). https://doi.org/10.1007/s00365-012-9170-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00365-012-9170-2
Keywords
- Spherical density estimation
- Local Hölder continuity
- Minimax bounds
- Needlets
- Thresholding
- Confidence intervals