Abstract
Statistical modeling plays a critical role in various scientific fields as it offers an understanding of how the response variable of interest is linked to a range of explanatory variables. However, selecting the best model beforehand is not easy. To address this issue, we conducted a study to explore the selection of an appropriate hyper-parameter for kernel semi-parametric regression. We used a quasi-oppositional learning pelican optimization algorithm technique to choose the smoothness parameter. Our simulation results showed that our proposed methodology outperformed its competitors in terms of mean squared error (MSE). Moreover, our suggested quasi-oppositional learning pelican optimization technique demonstrated superior performanceto the CV and GCV methods in computational time, as evidenced by experimental results and statistical analyses.
Recommended Citation
Yonis AL-Taie, Firas Ahmed; Qasim, Omar Saber; and Algamal, Zakariya Yahya
(2023)
"Kernel semi-parametric model improvement based on quasi-oppositional learning pelican optimization algorithm,"
Iraqi Journal for Computer Science and Mathematics: Vol. 4:
Iss.
2, Article 13.
DOI: https://doi.org/10.52866/ijcsm.2023.02.02.013
Available at:
https://ijcsm.researchcommons.org/ijcsm/vol4/iss2/13
