Multiple Regression Models Based on Gray Correlation Analysis and Factor Analysis Predict the Ownership of Civil Automobiles

Select Volume / Issue:

• Volume 7,Issue 3,May 2019

Year:

2019

Type of Publication:

Article

Keywords:

Grey Relational Analysis, Factor Analysis, Stepwise Regression

Authors:

Liu Yingjie

Journal:

IJISM

Volume:

7

Number:

3

Pages:

158-163

Month:

May

ISSN:

2347-9051

Abstract:

With the development of China's economy, the ownership of civil automobiles is increasing.The research and analysis of the civil automobile ownership is helpful for us to make a judgment on the trend of China's macro economy and provide reference for the development of the automobile industry. In this paper, grey correlation analysis was used to select the impact indicators, and factor analysis was used to construct the economic growth factor (Fac1), energy and environment factor (Fac2) and household consumption factor (Fac3). Then, stepwise regression was used to obtain the prediction model.The final model contains 8 economic indicators in total, including two factors, energy and environment factor and resident consumption factor, which greatly improves the disadvantage that regression analysis in other literatures can only get 2-3 explanatory variables in the end, also can provide a good explanation for the change of civilian automobile ownership.Compared with the time series method and grey prediction method, the method in this paper also ensures high accuracy. In this paper, a combined method, namely, selection of gray correlation analysis, factor analysis to construct impact factors, and regression analysis to obtain the final model, can not only predict China's civilian car ownership, but also explain the influencing factors.

Full text: IJISM_850_FINAL.pdf [Bibtex]