Abstract
We present several multivariate histogram density estimates that are universallyL 1-optimal to within a constant factor and an additive term\(O\left( {\sqrt {\log {n \mathord{\left/ {\vphantom {n n}} \right. \kern-\nulldelimiterspace} n}} } \right)\). The bin widths are chosen by the combinatorial method developed by the authors inCombinatorial Methods in Density Estimation (Springer-Verlag, 2001). The present paper solves a problem left open in that book.
Similar content being viewed by others
References
Abou-Jaoude, S. (1976a). Conditions nécessaires et suffisantes de convergencel 1 en probabilité de l'histogramme pour une densité.Annales de l'Institut Henri Poincaré, 12:213–231.
Abou-Jaoude, S. (1976b). La convergencel 1 etl ∞ de l'estimateur de la partition aleatoire pour une densité.Annales de l'Institut Henri Poincaré, 12:299–317.
Atilgan, T. (1990). On derivation and application of AIC as a data-based criterion for histograms.Communications in Statistics—Theory and Methods, 19:885–903.
Barron, A., Birgé, L., andMassart, P. (1999). Risk bounds for model selection via penalization.Probability Theory and Related Fields, 113:301–415.
Biau, G. andDevroye, L. (2002, to appear). On the risk of estimates for block decreasing densities.Journal of Multivariate Analysis.
Birgé, L. andRozenholc, Y. (2002). How many bins should be put in a regular histogram. Technical report.
Castellan, G. (2000). Sélection d'histogrammes ou de modèles exponentiels de polynomes par morceaux à l'aide d'un critère de type Akaike. Technical report.
Chen, X. R. andZhao, L. C. (1987). Almost sureL 1-norm convergence for data-based histogram density estimates.Journal of Multivariate Analysis, 21:179–188.
Devroye, L. (1987).A Course in Density Estimation. Birkhäuser-Verlag, Boston.
Devroye, L. andGyörfi, L. (1985).Nonparametric Density Estimation: The L 1 View. Wiley, New York.
Devroye, L. andLugosi, G. (1996). A universally acceptable smoothing factor for kernel density estimates.Annals of Statistics, 24:2499–2512.
Devroye, L. andLugosi, G. (1997). Nonasymptotic universal smoothing factors, kernel complexity and yatracos classes.Annals of Statistics, 25:2626–2637.
Devroye, L. andLugosi, G. (2001).Combinatorial Methods in Density Estimation. Springer-Verlag, New York.
Freedman, D. andDiaconis, P. (1981). On the histogram as a density estimator:l 2 theory.Zeitschrift für Wahrscheinlichkeitstheorie und verwandte Gebiete, 57:453–476.
Hall, P. (1990). Akaike's information criterion and Kullback-Leibler loss for histogram density estimation.Probability Theory and Related Fields, 85:449–467.
Hall, P. andHannan, E. (1988). On stochastic complexity and nonparametric density estimation.Biometrika, 75:705–714.
Kanazawa, Y. (1988). An optimal variable cell histogram.Communications in Statistics, part A: Theory and Methods, 17:1401–1422.
Kanazawa, Y. (1992). An optimal variable cell histogram based on the sample spacings.Annals of Statistics, 20:291–304.
Kanazawa, Y. (1993). Hellinger distance and Akaike's information criterion for the histogram.Statistics and Probability Letters, 17:293–298.
Kim, B. andRyzin, J. V. (1975). Uniform consistency of a histogram density estimator and modal estimation.Communications in Statistics, 4:303–315.
Kogure, A. (1987). Asymptotically optimal cells for a histogram.Annals of Statistics, 15:1023–1030.
Lecoutre, J.-P. (1985). Thel 2-optimal cell width for the histogram.Statistics and Probability Letters, 3:303–306.
Lugosi, G. andNobel, A. (1996). Consistency of data-driven histogram methods for density estimation and classification.Annals of Statistics, 24:687–706.
Rodriguez, C. andRyzin, J. V. (1985). Maximum entropy histograms.Statistics and Probability Letters, 3:117–120.
Rodriguez, C. andRyzin, J. V. (1986). Large sample properties of maximum entropy histograms.IEEE Transactions on Information Theory, IT-32:751–759.
Rudemo, M. (1982). Empirical choice of histograms and kernel density estimators.Scandinavian Journal of Statistics, 9:65–78.
Schläffli, L. (1950).Gesammelte Mathematische Abhandlungen. Birkhäuser-Verlag, Basel.
Scott, D. (1979). On optimal data-based histograms.Biometrika, 79:605–610.
Stone, C. J. (1985). An asymptotically optimal histogram selection rule. In L. L. Cam and R. A. Olshen, eds.,Proceedings of the Berkeley Conference in Honor of Jerzy Neyman and Jack Kiefer, Vol. II, pp. 513–520. Wadsworth, Belmont.
Taylor, C. (1987). Akaike's information criterion and the histogram.Biometrika, 74:636–639.
Vapnik, V. N. andChervonenkis, A. Y. (1971). On the uniform convergence of relative frequencies of events to their probabilities.Theory of Probability and its Applications, 16:264–280.
Wand, M. (1997). Data-based choice of histogram bin width.The American Statistician, 51:59–64.
Yu, B. andSpeed, T. (1990). Stochastic complexity and model selection II: Histograms. Technical report.
Yu, B. andSpeed, T. (1992). Data compression and histograms.Probability Theory and Related Fields, pp. 195–229.
Zhao, L. C., Krishnaiah, P. R., andChen, X. R. (1990). Almost sureL r-norm convergence for data-based histogram estimates.Theory of Probability and its Applications, 35:396–403.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Devroye, L., Lugosi, G. Bin width selection in multivariate histograms by the combinatorial method. Test 13, 129–145 (2004). https://doi.org/10.1007/BF02603004
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02603004
Key Words
- Multivariate density estimation
- nonparametric estimation
- variable histogram estimate
- bandwith selection