Prediction of Private Car Ownership in China Based on the Improved PCA-Logistic Model

DOI

10.2991/assehr.k.200331.074How to use a DOI?

Keywords

traffic engineering, private car ownership, improved principal component analysis (PCA), Logistic regression model, prediction

Abstract

The change of private car ownership is a long-term nonlinear process with complicated influencing factors and nonlinear relationship also exists among them, which requires reasonable and accurate methods for analysis and prediction. In order to solve these problems, the traditional principal component analysis (PCA) method is improved in this study. Considering the integrity of the data, eight representative main factors influencing China’s private car ownership are chosen. Firstly, the improved PCA method is used for “nonlinear” process of dimension reduction. Then, the Logistic regression model is applied to describe the relationship between private car ownership and the extracted principal components. The improved PCA-Logistic regression model is proposed finally. Compared with the results of the traditional PCA-Logistic method, it can be found that the improved PCA-Logistic model can effectively eliminate the nonlinear relationship between the data and it can change the quadratic nonlinear principle component regression curve obtained by the traditional method into a linear curve, which is more in line with the principle of PCA. The obtained nonlinear Logistic regression curve is in better agreement with the actual data, indicating that this method can evaluate the private car ownership in China more accurately. Based the model, the private car ownership in China from 2020 to 2024 have been predicted and the results show that China’s private car ownership may exceed 300 million in 2020 and exceed 400 million in 2024.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Bin Liu
AU  - Tianshu Zhao
AU  - Ranxia Zhang
PY  - 2020
DA  - 2020/04/02
TI  - Prediction of Private Car Ownership in China Based on the Improved PCA-Logistic Model
BT  - Proceedings of the 5th International Conference on Social Sciences and Economic Development (ICSSED 2020)
PB  - Atlantis Press
SP  - 353
EP  - 358
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.200331.074
DO  - 10.2991/assehr.k.200331.074
ID  - Liu2020
ER  -