The Annals of Statistics

Curve estimation when the design density is low

Peter Hall, J. S. Marron, M. H. Neumann, and D. M. Titterington

Full-text: Open access


In problems where a high-dimensional design is projected into a lower number of dimensions, the density of the new design is typically not bounded away from zero over its support, even if the original one was. Contexts where this problem arises include projection pursuit regression, estimation in single index models and application of the projection-slice method of Radon transform inversion. Theoretical work in these settings typically involves ignoring data toward the ends of the support of the projected design, but in practice that waste of information is not an attractive option. Motivated by these difficulties, we analyze the way in which local linear smoothing is affected by unboundedly sparse design and apply the conclusions of that study to develop empirical, adaptive bandwidth choice methods. Our results even add to knowledge in the familiar case of a design density that is bounded away from zero, where they provide adaptive bandwidth selectors that are optimal right to the ends of the design interval.

Article information

Ann. Statist., Volume 25, Number 2 (1997), 756-770.

First available in Project Euclid: 12 September 2002

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62G07: Density estimation
Secondary: 62G20: Asymptotic properties

Bandwidth choice design density kernel methods local adaptivity local inear smoothing projection pursuit single index model


Hall, Peter; Marron, J. S.; Neumann, M. H.; Titterington, D. M. Curve estimation when the design density is low. Ann. Statist. 25 (1997), no. 2, 756--770. doi:10.1214/aos/1031833672.

Export citation


  • Brillinger, D. R. (1983). A generalized linear model with "Gaussian" regressor variables. In A Festschrift for Erich Lehmann (P. J. Bickel, K. A. Doksum and J. L. Hodges, eds.) 97-114. Wadsworth, Belmont, CA.
  • Buckley, M. J. and Eagleson, G. K. (1989). A graphical method for estimating the residual variance in nonparametric regression. Biometrika 76 203-210.
  • Buckley, M. J., Eagleson, G. K. and Silverman, B. W. (1988). The estimation of residual variance in nonparametric regression. Biometrika 75 189-199.
  • Carter, C. K. and Eagleson, G. K. (1992). A comparison of variance estimators in nonparametric regression. J. Roy. Statist. Soc. Ser. B 54 773-780.
  • Cheng, M. Y., Fan, J. and Marron, J. S. (1996). Minimax efficiency of local poly nomial fit estimators at boundaries. Mimeo Series No. 2098, Institute of Statistics, Univ. North Carolina.
  • Cleveland, W. S. (1979). Robust locally weighted regression and smoothing scatterplots. J. Amer. Statist. Assoc. 74 829-836.
  • Cleveland, W. S. (1993). Visualizing Data. Hobart Press, Summit, NJ.
  • Cleveland, W. S. and Devlin, S. J. (1988). Locally weighted regression: an approach to regression analysis by local fitting. J. Amer. Statist. Assoc. 83 596-610.
  • Cleveland, W. S. and Grosse, E. H. (1991). Computational methods for local regression. Statist. and Computing 1 47-62.
  • Cleveland, W. S. and Loader, C. (1996). Smoothing by local regression: principles and methods. In Statistical Theory and Computational Aspects of Smoothing (W. H¨ardle and M. Schmlick, eds.) 10-49. physica, Heidelberg.
  • Fan, J. (1992). Design-adaptive nonparametric regression. J. Amer. Statist. Assoc. 87 998-1004.
  • Fan, J. (1993). Local linear smoothers and their minimax efficiency. Ann. Statist. 21 196-216.
  • Fan, J. and Gijbels, I. (1995). Data-driven bandwidth selection in local poly nomial fitting: variable bandwidth and spatial adaptation. J. Roy. Statist. Soc. Ser. B 57 371-394.
  • Friedman, J. H. and Stuetzle, W. (1981). Projection pursuit regression. J. Amer. Statist. Assoc. 76 817-823.
  • Gasser, T., Kneip, A. and K ¨ohler, W. (1991). A flexible and fast method for automatic smoothing. J. Amer. Statist. Assoc. 86 643-652.
  • Gasser, T., Sroka, L. and Jennen-Steinmetz, C. (1986). Residual variance and residual pattern in nonlinear regression. Biometrika 73 625-633.
  • Hall, P. and Hey de, C. C. (1980). Martingale Limit Theory and Its Application. Academic Press, New York.
  • Hall, P., Kay, J. W. and Titterington, D. M. (1990). Asy mptotically optimal difference-based estimation of variance in nonparametric regression. Biometrika 77 521-528.
  • Hall, P. and Marron, J. S. (1990). On variance estimation in nonparametric regression. Biometrika 77 415-419.
  • Hall, P., Marron, J. S., Neumann, M. H. and Titterington, D. M. (1995). On local linear smoothing when the design density is low. Research Report SRR027-95, Centre for Mathematics and Its Applications, Australian National Univ., Canberra.
  • H¨ardle, W., Hall, P. and Ichimura, H. (1993). Optimal smoothing in single-index models. Ann. Statist. 21 157-178.
  • Hastie, T. and Loader, C. (1993). Local regression: automatic kernel carpentry. Statist. Sci. 8 120-143.
  • Nicholls, D. F. and Pagan, A. R. (1985). Varying coefficient regression. In Handbook of Statistics 5 (E. J. Hannan, P. R. Krishnaiah and M. M. Rao, eds.) 413-449. North-Holland, Amsterdam.
  • Raj, B. and Ullah, A. (1981). Econometrics: A Varying Coefficients Approach. Croom-Helm, London.
  • Ruppert, D. and Wand, M. P. (1994). Multivariate locally weighted least squares regression. Ann. Statist. 22 1346-1370.
  • Seifert, B., Gasser, T. and Wolf, A. (1993). Nonparametric estimation of residual variance revisited. Biometrika 80 373-383.
  • Seifert, B. and Gasser, T. (1996). Finite sample analysis of local poly nomials: analysis and solutions. In Statistical Theory and Computational Aspects of Smoothing (W. H¨ardle and M. Schmlick, eds.) 50-79. physica, Heidelberg.