Bandwidth selection for density estimation by mixture with varying concentrations
DOI:
https://doi.org/10.17721/1812-5409.2025/2.6Keywords:
finite mixture model, varying concentrations, kernel density estimator, bandwidth selection, Silwerman’s rule of thumb, leave-one-out cross-validationAbstract
Finite mixture models arise in statistics of biological and medical data when the investigated subjects belong to sub-populations with different distributions of observed variable. In the model of mixture with varying concentrations (MVC) the concentrations of the mixture components can vary from observation to observation. We consider estimation of a probability density for a mixture component in MVC by a modification of the kernel density estimator (KDE). To apply KDE one needs to select a tuning parameter named the bandwidth. Two approaches to the bandwidth selection are considered. The first one is a modification of the Silverman‘s rule of thumb. The second one is a version of the leave-one-out cross-validation algorithm. We present results of simulation which show that both algorithms demonstrate similar behavior for nearly Gaussian densities and the cross-validation outperforms the Silverman‘s rule of thumb on highly non-Gaussian densities.
Pages of the article in the issue: 41 - 46
Language of the article: English
References
Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning data mining, inference, and prediction. Springer.
Maiboroda, R., Miroshnichenko, V., & Sugakova, O. (2022). Jackknife for nonlinear estimating equations. Modern Stochastics: Theory and Applications, 9(4), 377–399. https://doi.org/10.15559/22-VMSTA208
Maiboroda, R., Miroshnichenko, V., & Sugakova, O. (2024). Quantile estimators for regression errors in mixture models with varying
concentrations. Bulletin of Taras Shevchenko National University of Kyiv. Physical and Mathematical Sciences, 1(78), 45–50. https://doi.org/10.17721/1812-5409.2024/1.8
Maiboroda, R., & Sugakova, O. (2012). Statistics of mixtures with varying concentrations with application to dna microarray data analysis. Nonparametric statistics, 24(1), 201–215. https://doi.org/10.1080/10485252.2011.630076
Maiboroda, R., & Sugakova, O. (2020). Tests of hypotheses on quantiles of distributions of components in a mixture. Theory of Probability and Mathematical Statistics, 101, 179–191. https://doi.org/10.1090/tpms/1120
McLachlan, G., & Peel, D. (2000). Finite mixture models. Wiley. https://doi.org/10.1002/0471721182
Pidnebesna, A., Fajnerová, I., Horáček, J., & Hlinka, J. (2023). Mixture components inference for sparse regression: Introduction and application for estimation of neuronal signal from fmri bold. Applied Mathematical Modelling, 116, 735–748. https://doi.org/10.1016/j.apm.2022.11.034
Silverman, B. W. (2018). Density estimation for statistics and data analysis. Routledge.
Stone, C. (1984). An asymptotically optimal window selection rule for kernel density estimates. The Annals of Statistics, 12(4), 1285–1297. https://doi.org/10.1002/047172118210.1214/aos/1176346792
Sugakova, O. (1999). Asymptotics of a kernel estimate for the density of a distribution constructed from observations of a mixture with varying concentration. Theory of Probability and Mathematic Statistics, 59, 161–171.
Titterington, D., Smith, A., & Makov, O. (1985). Analysis of finite mixture distributions. Wiley.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Rostyslav Maiboroda, Olena Sugakova
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
