Converges in of nearest neighbor regression function estimate for strong mixing processes

Authors

  • Mohamed M. Deribati Tishreen University
  • Ahmad Younso Damascus University
  • Dema Al-shakh Tishreen University

Abstract

In this paper, we will study the issue of estimating regression function using k-nearest neighbors method (knn) for  mixing processes. We will extend the convergence in  for knn regression from independent case to the dependent case; In addition, we will conduct a simulation study using R software program to display the importance and influence of choosing number of neighbors (k) and the sample size (n) on behavior of the estimator. For this purpose, the mean squares error criterion (MSE) was used. The high variability of the MSE results shows that knn estimators are very sensitive to the choice of the number of neighbors. More results show that the higher value of n, the more accurate and effective the estimator.

Author Biographies

Mohamed M. Deribati, Tishreen University

Associate Professor, Depart. Of Mathematical Statistics, Faculty of Science

 

Ahmad Younso, Damascus University

Associate Professor, Depart. Of Mathematical Statistics, Faculty of Science

 

Dema Al-shakh, Tishreen University

Postgraduate student, Depart. Of Mathematical Statistics

 

Published

2022-01-15

How to Cite

1.
Deribati MM, Younso A, Al-shakh D. Converges in of nearest neighbor regression function estimate for strong mixing processes. TUJ-BA [Internet]. 2022Jan.15 [cited 2022May16];43(6):59-71. Available from: http://journal.tishreen.edu.sy/index.php/bassnc/article/view/11509