Adaptive estimation of a quadratic functional of a density by model selection

Laurent, Béatrice

ESAIM: Probability and Statistics, Tome 9 (2005), pp. 1-18.

We consider the problem of estimating the integral of the square of a density $f$ from the observation of a $n$ sample. Our method to estimate ${\int }_{ℝ}{f}^2\left(x\right)\mathrm{d}x$ is based on model selection via some penalized criterion. We prove that our estimator achieves the adaptive rates established by Efroimovich and Low on classes of smooth functions. A key point of the proof is an exponential inequality for $U$-statistics of order 2 due to Houdré and Reynaud.

Classification : 62G05, 62G20, 62J02
Mots clés : adaptive estimation, quadratic functionals, model selection, Besov bodies, efficient estimation

[1] P. Bickel and Y. Ritov, Estimating integrated squared density derivatives: sharp best order of convergence estimates. Sankhya Ser. A. 50 (1989) 381-393. | Zbl

[2] L. Birgé and P. Massart, Estimation of integral functionals of a density. Ann. Statist. 23 (1995) 11-29. | Zbl

[3] L. Birgé and P. Massart, Minimum contrast estimators on sieves: exponential bounds and rates of convergence. Bernoulli 4 (1998) 329-375. | Zbl

[4] L. Birgé and Y. Rozenholc, How many bins should be put in a regular histogram. Technical Report Université Paris 6 et 7 (2002).

[5] J. Bretagnolle, A new large deviation inequality for $U$-statistics of order 2. ESAIM: PS 3 (1999) 151-162. | Numdam | Zbl

[6] D. Donoho and M. Nussbaum, Minimax quadratic estimation of a quadratic functional. J. Complexity 6 (1990) 290-323. | Zbl

[7] S. Efroïmovich and M. Low, On Bickel and Ritov's conjecture about adaptive estimation of the integral of the square of density derivatives. Ann. Statist. 24 (1996) 682-686. | Zbl

[8] S. Efroïmovich and M. Low, On optimal adaptive estimation of a quadratic functional. Ann. Statist. 24 (1996) 1106-1125. | Zbl

[9] M. Fromont and B. Laurent, Adaptive goodness-of-fit tests in a density model. Technical report. Université Paris 11 (2003). | Zbl

[10] G. Gayraud and K. Tribouley, Wavelet methods to estimate an integrated quadratic functional: Adaptivity and asymptotic law. Statist. Probab. Lett. 44 (1999) 109-122. | Zbl

[11] E. Giné, R. Latala and J. Zinn, Exponential and moment inequalities for $U$-statistics. High Dimensional Probability 2, Progress in Probability 47 (2000) 13-38. | Zbl

[12] W. Hardle, G. Kerkyacharian, D. Picard, A. Tsybakov, Wavelets, Approximations and statistical applications. Lect. Notes Stat. 129 (1998). | MR | Zbl

[13] C. Houdré and P. Reynaud-Bouret, Exponential inequalities for U-statistics of order two with constants, in Euroconference on Stochastic inequalities and applications. Barcelona. Birkhauser (2002). | Zbl

[14] I.A. Ibragimov, A. Nemirovski and R.Z. Hasminskii, Some problems on nonparametric estimation in Gaussian white noise. Theory Probab. Appl. 31 (1986) 391-406. | Zbl

[15] I. Johnstone, Chi-square oracle inequalities. State of the art in probability and statistics (Leiden 1999) - IMS Lecture Notes Monogr. Ser., 36. Inst. Math. Statist., Beachwood, OH (1999) 399-418.

[16] B. Laurent, Efficient estimation of integral functionals of a density. Ann. Statist. 24 (1996) 659-681. | Zbl

[17] B. Laurent, Estimation of integral functionals of a density and its derivatives. Bernoulli 3 (1997) 181-211. | Zbl

[18] B. Laurent and P. Massart, Adaptive estimation of a quadratic functional by model selection. Ann. Statist. 28 (2000) 1302-1338. | Zbl