Bandwidth Selection in Multivariate Nadaraya-Watson Estimator based on Meta-Heuristic Optimization Algorithms: A Simulation Study

Abstract

The smoothing parameter has a complete bearing on curve estimation in the context of kernel nonparametric regression.Meta-heuristic algorithm implementation has grown in popularity among researchers.In this study, a multivariate Nadaraya-Watson kernel nonparametric regression bandwidth matrix selection approach based on a pigeon optimization algorithm is given.The suggested approach will effectively assist in identifying the appropriate bandwidth matrix with a strong forecast.The proposed approach is contrasted with two well-known approaches.The thorough demonstration of the suggested method's superiority in terms of prediction ability is provided by the experimental results.