On the asymptotic properties of a simple estimate of the Mode

Christophe Abraham; Gérard Biau; Benoît Cadre

ESAIM: Probability and Statistics (2004)

  • Volume: 8, page 1-11
  • ISSN: 1292-8100

Abstract

top
We consider an estimate of the mode θ of a multivariate probability density f with support in d using a kernel estimate f n drawn from a sample X 1 , , X n . The estimate θ n is defined as any x in { X 1 , , X n } such that f n ( x ) = max i = 1 , , n f n ( X i ) . It is shown that θ n behaves asymptotically as any maximizer θ ^ n of f n . More precisely, we prove that for any sequence ( r n ) n 1 of positive real numbers such that r n and r n d log n / n 0 , one has r n θ n - θ ^ n 0 in probability. The asymptotic normality of θ n follows without further work.

How to cite

top

Abraham, Christophe, Biau, Gérard, and Cadre, Benoît. "On the asymptotic properties of a simple estimate of the Mode." ESAIM: Probability and Statistics 8 (2004): 1-11. <http://eudml.org/doc/245948>.

@article{Abraham2004,
abstract = {We consider an estimate of the mode $\theta $ of a multivariate probability density $f$ with support in $\mathbb \{R\}^d$ using a kernel estimate $f_n$ drawn from a sample $X_1, \hdots , X_n$. The estimate $\theta _n$ is defined as any $x$ in $\lbrace X_1, \hdots , X_n\rbrace $ such that $f_n(x)=\max _\{i=1, \hdots ,n\} f_n(X_i)$. It is shown that $\theta _n$ behaves asymptotically as any maximizer $\{\hat\{\theta \}\}_n$ of $f_n$. More precisely, we prove that for any sequence $(r_n)_\{n\ge 1\}$ of positive real numbers such that $r_n\rightarrow \infty $ and $r_n^d\log n/n\rightarrow 0$, one has $r_n\,\Vert \theta _n-\{\hat\{\theta \}\}_n\Vert \rightarrow 0$ in probability. The asymptotic normality of $\theta _n$ follows without further work.},
author = {Abraham, Christophe, Biau, Gérard, Cadre, Benoît},
journal = {ESAIM: Probability and Statistics},
keywords = {multivariate probability density; mode; kernel estimate; central limit theorem},
language = {eng},
pages = {1-11},
publisher = {EDP-Sciences},
title = {On the asymptotic properties of a simple estimate of the Mode},
url = {http://eudml.org/doc/245948},
volume = {8},
year = {2004},
}

TY - JOUR
AU - Abraham, Christophe
AU - Biau, Gérard
AU - Cadre, Benoît
TI - On the asymptotic properties of a simple estimate of the Mode
JO - ESAIM: Probability and Statistics
PY - 2004
PB - EDP-Sciences
VL - 8
SP - 1
EP - 11
AB - We consider an estimate of the mode $\theta $ of a multivariate probability density $f$ with support in $\mathbb {R}^d$ using a kernel estimate $f_n$ drawn from a sample $X_1, \hdots , X_n$. The estimate $\theta _n$ is defined as any $x$ in $\lbrace X_1, \hdots , X_n\rbrace $ such that $f_n(x)=\max _{i=1, \hdots ,n} f_n(X_i)$. It is shown that $\theta _n$ behaves asymptotically as any maximizer ${\hat{\theta }}_n$ of $f_n$. More precisely, we prove that for any sequence $(r_n)_{n\ge 1}$ of positive real numbers such that $r_n\rightarrow \infty $ and $r_n^d\log n/n\rightarrow 0$, one has $r_n\,\Vert \theta _n-{\hat{\theta }}_n\Vert \rightarrow 0$ in probability. The asymptotic normality of $\theta _n$ follows without further work.
LA - eng
KW - multivariate probability density; mode; kernel estimate; central limit theorem
UR - http://eudml.org/doc/245948
ER -

References

top
  1. [1] C. Abraham, G. Biau and B. Cadre, Simple estimation of the mode of a multivariate density. Canadian J. Statist. 31 (2003) 23-34. Zbl1035.62046MR1985502
  2. [2] L. Devroye, Recursive estimation of the mode of a multivariate density. Canadian J. Statist. 7 (1979) 159-167. Zbl0444.60023MR570537
  3. [3] L. Devroye, A Course in Density Estimation. Birkhäuser, Boston (1987). Zbl0617.62043MR891874
  4. [4] W.F. Eddy, Optimum kernel estimates of the mode. Ann. Statist. 8 (1980) 870-882. Zbl0438.62027MR572631
  5. [5] V.D. Konakov, On asymptotic normality of the sample mode of multivariate distributions. Theory Probab. Appl. 18 (1973) 836-842. Zbl0325.62033MR336874
  6. [6] J. Leclerc and D. Pierre-Loti-Viaud, Vitesse de convergence presque sûre de l’estimateur à noyau du mode. C. R. Acad. Sci. Paris 331 (2000) 637-640. Zbl0961.62043
  7. [7] A. Mokkadem and M. Pelletier, A law of the iterated logarithm for the kernel mode estimator, ESAIM: Probab. Statist. 7 (2003) 1-21. Zbl1013.62032MR1956072
  8. [8] E. Parzen, On estimation of a probability density function and mode. Ann. Math. Statist. 33 (1962) 1065-1076. Zbl0116.11302MR143282
  9. [9] D. Pollard, Convergence of Stochastic Processes. Springer–Verlag, New York (1984). Zbl0544.60045
  10. [10] J.P. Romano, On weak convergence and optimality of kernel density estimates of the mode. Ann. Statist. 16 (1988) 629-647. Zbl0658.62053MR947566
  11. [11] M. Rosenblatt, Remarks on some nonparametric estimates of a density function. Ann. Math. Statist. 27 (1956) 832-837. Zbl0073.14602MR79873
  12. [12] T.W. Sager, Estimating modes and isopleths. Comm. Statist. – Theory Methods 12 (1983) 529-557. Zbl0513.62049
  13. [13] M. Samanta, Nonparametric estimation of the mode of a multivariate density. South African Statist. J. 7 (1973) 109-117. Zbl0268.62015MR331618
  14. [14] B. Silverman, Weak and strong uniform consistency of the kernel estimate of a density and its derivatives. Ann. Statist. 6 (1978) 177-184. Zbl0376.62024MR471166
  15. [15] P. Vieu, A note on density mode estimation. Statist. Probab. Lett. 26 (1996) 297-307. Zbl0847.62024MR1393913

NotesEmbed ?

top

You must be logged in to post comments.