A Convenient Ready-to-Use Algorithm for a Conditional Quantile Estimator

Authors

Piotr Kulczycki, Polish Academy of Sciences, Systems Research Institute, Center of Information Technology for Data Analysis Methods, Newelska 6, 01-447 Warsaw, Poland\\ AGH University of Science and Technology, Faculty of Physics and Applied Computer Science, Division of Information Technology and Biometrics, Mickiewicza 30, 30-059 Cracow, PolandFollow
Małgorzata Charytanowicz, Polish Academy of Sciences, Systems Research Institute, Center of Information Technology for Data Analysis Methods, Newelska 6, 01-447 Warsaw, Poland\\ John Paul II Catholic University of Lublin, Institute ofMathematics and Computer Science, Konstantyn´ow 1H, 20-708 Lublin, PolandFollow
Antoni Leon Dawidowicz, Jagiellonian University, Institute of Mathematics, Łojasiewicza 6, 30-348 Cracow, PolandFollow

Author Country (or Countries)

Poland

Abstract

This paper contains a complete procedure for calculating the value of a conditional quantile estimator. The concept is based on the nonparametric kernel estimator method, which frees the algorithm from the random variables’ distributions. The procedure was worked out in a ready-to-use form – specific formulas for functions and the parameter used were given. The practical implementation of this method is very simple, and its computational complexity is linear with respect to random sample size as well as the dimension of conditioning variable. Thanks to a clear, near intuitive interpretation it can easily be modified or generalized depending on the individual needs of atypical applications. In particular, conditioning variables can be taken into account, not only continuous (real), but also binary, discrete and categorized, or any of their combinations.

Recommended Citation

Kulczycki, Piotr; Charytanowicz, Małgorzata; and Leon Dawidowicz, Antoni (2015) "A Convenient Ready-to-Use Algorithm for a Conditional Quantile Estimator," Applied Mathematics & Information Sciences: Vol. 09: Iss. 2, Article 33.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol09/iss2/33