ИСТИНА |
Войти в систему Регистрация |
|
ИСТИНА ИНХС РАН |
||
Along with a brief survey of various methods employed in the feature selection theory (see, e.g., [1]) we develop the recent papers [2] and [3] to study statistical estimation of mutual information and other divergences. Such estimates are used for identification of relevant factors having impact on a random response. This research direction is very important, e.g., for analysis of biological and medical data. Theoretical results describing the asymptotic behavior of statistics under consideration are supplemented with computer simulations. Special attention is paid to the so-called mixed models comprising the widely used logistic regression. REFERENCES [1] Bolon-Canedo, V., Alonso-Betanzos, A. Recent Advances in Ensembles for Feature Selection. Springer, 2018. [2] Bulinski, A., Dimitrov, D. Statistical estimation of the Shannon entropy. Acta Mathematica Sinica. English Series. Published online: September 7 (2018), p.1-28. DOI: https://doi.org/10.1007/s10114-018-7440-z. [3] Bulinski, A., Kozhevin, A. Statistical Estimation of Conditional Shannon Entropy. ESAIM: Probability and Statistics. Published online: November 28 (2018), p.1-35.