2022-05

CMES - International Conference on Computational Mathematics and Engineering Sciences

Abstract--The point of this research is to compare K-Nearest Neighbor Model with Kernel Model in nonparametric functional data with conditional supposition in the case of response as a scalar variable while the function is covariates. We utilize the form of the Nadaraya-Watson estimator method for prediction with two kinds of semi-metrics: semi-metric built on the second derivatives and the second based on Functional Principle Component Analysis. The results get from K-Nearest Neighbor method is more accurate than the outcomes get from kernel model. The achievement of this research is then assessed by computing mean square errors. The method is illustrated by some applications.

2018-07

Postgraduate Research Conference

In this chapter, we propose a new model for nonparametric functional regression analysis with conditional expectation in the context of response as a multivariate variable while the covariates take values in some innite dimensional space. We use the formula of the Nadaraya-Watson estimator (k-nearest neighbour (KNN)) for prediction with two kinds of semi-metrics as a measure of proximity between curves(semi-metric based on the derivatives and semi-metric built on the Functional Principal Component analysis). The present model is more convenient for the prediction of the components of a vector of random variables together rather than for predicting each of them separately because of saving time and least cost. The performance of this model is then evaluated by calculating mean square prediction errors which are then compared to the independent response model. The model is illustrated by both a simulation study and analysis of two real data sets from functional data analysis (Spectrometric Data and Canadian Weather Stations). The chapter is organized as follows. Section 3:1 introduces the multivariate response variable method with K-Nearest Neighbour estimator. Section 3:2 contains the simulation study. Two examples of treatment of real data are presented in Section 3:3. Finally, some concluding remarks will be given in Section 3:4.

2017-03

The Role of the Kurds in the Middle East and beyond: Regional and International Interactions

only I attended

2016-06

Kurdish Studies Summer School

only I attended

2016-04

“Hilbert’s Sixth Problem”.

Only I attended