zbMATH — the first resource for mathematics

Two-component mixtures with independent coordinates as conditional mixtures: nonparametric identification and estimation. (English) Zbl 1337.62056
Summary: We show how the multivariate two-component mixtures with independent coordinates in each component by P. Hall and X.-H. Zhou [Ann. Stat. 31, No. 1, 201–224 (2003; Zbl 1018.62021)] can be studied within the framework of conditional mixtures as recently introduced by M. Henry et al. [“Identifying finite mixtures in econometric models”, Cowles Foundation Discussion Paper Series 1767 (2010), http://cowles.yale.edu/sites/default/files/files/pub/d17/d1767.pdf]. Here, the conditional distribution of the random variable \(Y\) given the vector of regressors \(Z\) can be expressed as a two-component mixture, where only the mixture weights depend on the covariates. Under appropriate tail conditions on the characteristic functions and the distribution functions of the mixture components, which allow for flexible location-scale type mixtures, we show identification and provide asymptotically normal estimators. The main application for our results are bivariate two-component mixtures with independent coordinates, the case not previously covered by Hall and Zhou [loc. cit.]. In a simulation study we investigate the finite-sample performance of the proposed methods. The main new technical ingredient is the estimation of limits of quotients of two characteristic functions in the tails from independent samples, which might be of some independent interest.

62G05 Nonparametric estimation
62H12 Estimation in multivariate analysis
62G20 Asymptotic properties of nonparametric inference
Full Text: DOI Euclid
[1] Bordes, L., Mottelet, S. and Vandekerkhove, P. (2006). Semiparametric Estimation of a Two-component Mixture Model. Ann. Statist. 34 1204-1232. · Zbl 1112.62029
[2] Bordes, L. and Vandekerkhove, P. (2010). Semiparametric two-component mixture model when a component is known: an asymptotically normal estimator. Mathematical Methods of Statistics 19 22-41. · Zbl 1282.62068
[3] Csörgő, S. (1981). Limit behaviour of the Empirical Characteristic Function. Ann. Probab. 9 130-144. · Zbl 0453.60025
[4] Hall, P. and Zhou, X. (2003). Nonparametric Estimation of Component Distributions in a Multivariate Mixture. Ann. Statist. 31 201-224. · Zbl 1018.62021
[5] Henry, M., Kitamura, Y. and Salanié, B. (2010). Identifying Finite Mixtures in Econometric Models. Cowles Foundation Discussion Paper Series 1767 .
[6] Hohmann, D. and Holzmann, H. (2012). Semiparametric location mixtures with distinct components. Statistics . · Zbl 1440.62108
[7] Hohmann, D. and Holzmann, H. (2013). Two-component mixtures with independent coordinates as conditional mixtures: Nonparametric identification and estimation. Technical Report, Marburg University . · Zbl 1337.62056
[8] Hunter, D., Wang, S. and Hettmansperger, T. (2007). Inference for Mixtures of symmetric distribution. Ann. Statist. 35 224-251. · Zbl 1114.62035
[9] Kasahara, H. and Shimotsu, K. (2009). Nonparametric Identification of Finite Mixture Models of Dynamic Discrete Choices. Econometrica vol. 77 135-175. · Zbl 1160.91323
[10] Kawata, T. (1972). Fourier Analysis in Probability Theory . Probability and Mathematical Statistics . Academic Press, New York. · Zbl 0271.60022
[11] Kitamura, Y. (2004). Nonparametric Identifiability of Finite Mixtures Unpublished manuscript, Yale University, Department of Economics.
[12] McLachlan, G. and Peel, D. (2000). Finite Mixture Models . Wiley series in probability and statistics . John Wiley & Sons, New York. · Zbl 0963.62061
[13] Talagrand, M. (1996). Lower Classes for Fractional Brownian Motion. J. Theor. Probab. 9 191-213. · Zbl 0840.60076
[14] van der Vaart, A. and Wellner, J. (2000). Weak Convergence and Empirical Processes . Springer Series in Statistics . Springer, New York. · Zbl 0967.60037
[15] Vandekerkhove, P. (2010). Estimation of a semiparametric contaminated regression model. · Zbl 1282.62068
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.