Electronic Journal of Statistics

Revisiting the Hodges-Lehmann estimator in a location mixture model: Is asymptotic normality good enough?

Fadoua Balabdaoui

Full-text: Open access

Abstract

Location mixture models, resulting in shifting a common distribution with some probability, have been widely used to account for existence of clusters in the data. Assuming only symmetry of this common distribution allows for great flexibility, especially when the traditional normality assumption is violated. This semi-parametric model has been studied in several papers, where the mixture parameters are first estimated before constructing an estimator for the non-parametric component. The plug-in method suggested by Hunter et al. (2007) has the merit to be easily implementable and fast to compute. However, no result is available on the limit distribution of the obtained estimator, hindering for instance construction of asymptotic confidence intervals. In this paper, we give sufficient conditions on the symmetric distribution for asymptotic normality to hold. In case the symmetric distribution admits a log-concave density, our assumptions are automatically satisfied. The obtained result has to be used with caution in case the mixture location are too close or the mixing probability is close to $0$ or $1$. Three examples are considered where we show that the estimator is not to be advocated when the mixture components are not well separated.

Article information

Source
Electron. J. Statist., Volume 11, Number 2 (2017), 4563-4595.

Dates
Received: December 2016
First available in Project Euclid: 17 November 2017

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1510887946

Digital Object Identifier
doi:10.1214/17-EJS1311

Mathematical Reviews number (MathSciNet)
MR3724489

Zentralblatt MATH identifier
1380.62230

Rights
Creative Commons Attribution 4.0 International License.

Citation

Balabdaoui, Fadoua. Revisiting the Hodges-Lehmann estimator in a location mixture model: Is asymptotic normality good enough?. Electron. J. Statist. 11 (2017), no. 2, 4563--4595. doi:10.1214/17-EJS1311. https://projecteuclid.org/euclid.ejs/1510887946


Export citation

References

  • Balabdaoui, F. (2017). Supplement A: Proofs of Theorem 2.6 and 2.7 and some useful formulae., Electronic Journal of Statistics. doi: http://dx.doi.org/10.1214/17-EJS1311SUPP.
  • Balabdaoui, F. and Doss, C. (2016). Inference for a two-component mixture of symmetric distributions under log-concavity., Bernoulli to appear.
  • Bordes, L., Mottelet, S. and Vandekerkhove, P. (2006). Semiparametric estimation of a two-component mixture model., Ann. Statist. 34 1204‒1232.
  • Butucea, C. and Vandekerkhove, P. (2014). Semiparametric mixtures of symmetric distributions., The Scandinavian Journal of Statistics 41 227‒239.
  • Chang, G. T. and Walther, G. (2007). Clustering with mixtures of log-concave distributions., Computational Statistics & Data Analysis 51 6242‒6251.
  • Chee, C.-S. and Wang, Y. (2013). Estimation of finite mixtures with symmetric components., Stat. Comput. 23 233‒249. URL http://dx.doi.org/10.1007/s11222-011-9305-5
  • Cule, M., Samworth, R. and Stewart, M. (2010). Maximum likelihood estimation of a multidimensional log-concave density., J. R. Stat. Soc. Ser. B Stat. Methodol. 72 545‒607.
  • Finner, H. and Roters, M. (1993). Distribution functions and log-concavity., Comm. Statist. Theory Methods 22 2381‒2396. URL http://dx.doi.org/10.1080/03610929308831156
  • Hunter, D. R., Wang, S. and Hettmansperger, T. P. (2007). Inference for mixtures of symmetric distributions., Ann. Statist. 35 224‒251. URL http://dx.doi.org/10.1214/009053606000001118
  • Ma, Y. and Yao, W. (2015). Flexible estimation of a semiparametric two-component mixture model with one parametric component., Electron. J. Stat. 9 444‒474. URL http://dx.doi.org/10.1214/15-EJS1008
  • Schoenberg, I. J. (1951). On Pólya frequency functions. I. The totally positive functions and their Laplace transforms., J. Analyse Math. 1 331‒374.
  • van der Vaart, A. W. and Wellner, J. A. (1996)., Weak convergence and empirical processes. Springer Series in Statistics, Springer-Verlag, New York. With applications to statistics.

Supplemental materials

  • Supplement A: Proofs of Theorem 2.6 and 2.7 and some useful formulae. In this supplementary file we provide proofs of Theorem 2.6 and 2.7 describing the weak limiting distribution of the estimator of Hunter et al. (2007). Some formulae used in the derivation of this limit are also given.