• ISSN: 2010-023X (Print)
    • Abbreviated Title: Int. J. Trade, Economics and Financ.
    • Frequency: Quaterly
    • DOI: 10.18178/IJTEF
    • Editor-in-Chief: Prof.Tung-Zong (Donald) Chang
    • Managing Editor: Ms. Inez. Chan

    • Abstracting/ Indexing:  Crossref, CNKI, EBSCO

    • Article Processing Charge (APC): 500 USD

    • E-mail: ijtef.editorial.office@gmail.com

HOME > Archive > 2022 > Volume 13 Number 2 (Apr. 2022) >
IJTEF 2022 Vol.13(2): 28-35 ISSN: 2010-023X
DOI: 10.18178/ijtef.2022.13.2.719

Feature Importance Analysis in Global Manufacturing Industry

Kenji Yamaguchi

Abstract—SHAP is a measurement based on Shapley values and has been used widely in machine-learning regressions to interpret the feature importance. I conducted the feature importance analysis by the SHAP values in the global manufacturing industry. The target fields are automakers and electronic companies. I found the interesting attribute of Shapley values through the regression analysis. In general, the predictor variable values of companies forge no linear relationship to the target values such as a profit ratio. However, after making the SHAP values for each predictor, the scattering plot between the SHAP values and the target values clarifies the linear relationship between them. I verified the linear relationship on both automakers and electronic companies. The insight of the linearity is presented in this paper. Each company has a different behavioral structure specific to the company. The SHAP value extracts the company’s behavioral structure through the characteristic function.
In addition, to make the regression results more precise and avoid effects by the multi-collinearity, I conducted a PCA (Principal Component Analysis). From the 3D scatter plot of the PCA of SHAP values, I verified the linear relationship as I expected and could identify the latent semantics of the PC1 and PC2 as a profitability related factor and an operation management relation factor.

Index Terms—Shapley values, characteristic function, company performance measurement, machine learning, regression, global manufacturing.

K. Yamaguchi is with Science & Education Center (SEC), Ochanomizu University, Japan (e-mail: yamaguchi.kenji@ocha.ac.jp).

[PDF]

Cite: Kenji Yamaguchi, "Feature Importance Analysis in Global Manufacturing Industry," International Journal of Trade, Economics and Finance vol.13, no.2, pp. 28-35, 2022.

Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).

PREVIOUS PAPER
First page
NEXT PAPER
A Study on Analytical and Numerical Solutions of Three Types of Black-Scholes Models

Copyright © 2008-2024. International Journal of Trade, Economics and Finance. All rights reserved.
E-mail: ijtef.editorial.office@gmail.com