A working likelihood approach to support vector regression with a data-driven insensitivity parameter

TL;DR

This paper introduces a data-driven method for selecting the insensitivity parameter in support vector regression, improving prediction accuracy and computational efficiency by minimizing a likelihood-based loss function.

Contribution

It proposes a novel likelihood-based approach to automatically determine the insensitivity parameter in support vector regression, enhancing model performance and standardizing samples.

Findings

Outperforms existing methods in simulations and real data tests.

Achieves lower computational costs compared to traditional approaches.

Demonstrates robustness across different noise distributions.

Abstract

The insensitive parameter in support vector regression determines the set of support vectors that greatly impacts the prediction. A data-driven approach is proposed to determine an approximate value for this insensitive parameter by minimizing a generalized loss function originating from the likelihood principle. This data-driven support vector regression also statistically standardizes samples using the scale of noises. Nonlinear and linear numerical simulations with three types of noises ($\epsilon$-Laplacian distribution, normal distribution, and uniform distribution), and in addition, five real benchmark data sets, are used to test the capacity of the proposed method. Based on all of the simulations and the five case studies, the proposed support vector regression using a working likelihood, data-driven insensitive parameter is superior and has lower computational costs.