The Annals of Statistics

A Canonical Process for Estimation of Convex Functions: The "Invelope" of Integrated Brownian Motion $+t^4$

Piet Groeneboom, Geurt Jongbloed, and Jon A. Wellner

Full-text: Open access


A process associated with integrated Brownian motion is introduced that characterizes the limit behavior of nonparametric least squares and maximum likelihood estimators of convex functions and convex densities, respectively. We call this process “the invelope” and show that it is an almost surely uniquely defined function of integrated Brownian motion. Its role is comparable to the role of the greatest convex minorant of Brownian motion plus a parabolic drift in the problem of estimating monotone functions. An iterative cubic spline algorithm is introduced that solves the constrained least squares problem in the limit situation and some results, obtained by applying this algorithm, are shown to illustrate the theory.

Article information

Ann. Statist., Volume 29, Number 6 (2001), 1620-1652.

First available in Project Euclid: 5 March 2002

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62G05: Estimation
Secondary: 60G15: Gaussian processes 62E20: Asymptotic distribution theory

Convex function estimation Gaussian process integrated Brownian motion least squares


Groeneboom, Piet; Jongbloed, Geurt; Wellner, Jon A. A Canonical Process for Estimation of Convex Functions: The "Invelope" of Integrated Brownian Motion $+t^4$. Ann. Statist. 29 (2001), no. 6, 1620--1652. doi:10.1214/aos/1015345957.

Export citation


  • Brunk, H. D. (1970). Estimation of isotonic regression. In Nonparametric Techniques in Statistical Inference (M. L. Puri, ed.). Cambridge Univ. Press.
  • Dieudonn´e, J. (1969). Foundations of Modern Analysis. Academic Press, New York.
  • Groeneboom, P. (1985). Estimating a monotone density. In Proceedings of the Berkeley Conference in Honor of Jerzy Neyman and Jack Kiefer (L. M. LeCam and R. A. Olshen, eds.) 2 Wadsworth, Hayward, CA.
  • Groeneboom, P. (1988). Brownian motion with a parabolic drift and Airy functions. Probab. Theory Related Fields 81 79-109.
  • Groeneboom, P., Jongbloed, G. and Wellner, J. A. (1999). Integrated Brownian motion conditioned to be positive. Ann. Probab. 27 1283-1303. Groeneboom, P., Jongbloed, G. and Wellner, J. A. (2001a). Estimation of convex functions: characterizations and asymptotic theory. Ann. Statist. 29 1653-1698. Groeneboom, P., Jongbloed, G. and Wellner, J. A. (2001b). Computation of nonparametric function estimators via vertex direction algorithms. Unpublished manuscript.
  • Groeneboom, P. and Wellner, J. A. (2001). Computing Chernoff's distribution. J. Comput. Graph. Statist. 10 388-400.
  • Jongbloed, G. (1995). Three Statistical Inverse Problems. Ph.D. dissertation, Delft Univ. Technology.
  • Mammen, E. (1991). Nonparametric regression under qualitative smoothness assumptions. Ann. Statist. 19 741-759.
  • Meyer, M. C. (1997). Shape restricted inference with applications to nonparametric regression, smooth nonparametric function estimation, and density estimation. Ph.D. dissertation, Dept. Statistics, Univ. Michigan, Ann Arbor.
  • Prakasa Rao, B. L. S. (1969). Estimation of a unimodal density. Sankhy¯a. Ser. A 31 23-36.
  • Press, W. H., Teukolsky, S. A., Vetterling, W. T. and Flannery, B. P. (1992). Numerical Recipes in C. Cambridge Univ. Press.
  • Rogers, L. C. G. and Williams, D. (1994). Diffusions, Markov Processes and Martingales 1, 2nd ed. Wiley, New York.
  • Sinai, Y. G. (1992). Statistics of shocks in solutions of inviscid Burgers equation. Comm. Math. Phys. 148 601-621.
  • Wang, Y. (1994). The limiting distribution in concave regression. Preprint, Univ. Missouri, Columbia.
  • Wright, S. J. (1997). Primal-Dual Interior Point Methods. SIAM, Philadelphia.