Journal
SOFT COMPUTING
Volume 24, Issue 6, Pages 4259-4268Publisher
SPRINGER
DOI: 10.1007/s00500-019-04191-0
Keywords
Grey prediction; Time series; Multi-criteria decision making; Feature selection; Bankruptcy prediction
Categories
Funding
- Ministry of Science and Technology, Taiwan [MOST 106-2410-H-033-006-MY2]
Ask authors/readers for more resources
Regarding bankruptcy prediction as a kind of grey system problem, this study aims to develop multivariate grey prediction models based on the most representative GM(1, N) for bankruptcy prediction. There are several distinctive features of the proposed grey prediction model. First, to improve the prediction performance of the GM(1, N), grey relational analysis is used to sift relevant features that have the strongest relationship with the class feature. Next, the proposed model effectively extends the multivariate grey prediction model for time series to bankruptcy prediction irrespective of time series. It turns out that the proposed model uses the genetic algorithms to avoid indexing by time and using the ordinary least squares with statistical assumptions for the traditional GM(1, N). The empirical results obtained from the financial data of Taiwanese firms in the information and technology industry demonstrated that the proposed prediction model performs well compared with other GM(1, N) variants considered.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available