Financial Distress Study Based on PSO K-Means Clustering Algorithm and Rough Set Theory

Abstract:

The traditional financial distress method normally divided samples into two categories by healthy and bankruptcy. And the financial indicators are typically chosen without using a systematic and reasonable theory. To be more realistic, this paper selected all the companies in a certain industry as the research objects. Twenty-one financial indicators were primarily chosen as the condition attributes, reduction set was obtained by matrix reduction identification based on rough set theory. Then PSO-based clustering algorithm K-means was used to divide subjects into 5 categories of different financial status. The decision-making table was formed with the reduction set using the classification as a decision attribute. Finally, we tested the reasonableness of the classification and generated early warning rules together with rough set theory to evaluate the financial status of listed companies. The results showed that PSO-based K-means algorithm was able to reasonably classify companies, at the same time to overcome the subjective impacts in the artificial measure of financial crisis level. Data generated using this method agreed with the rough set theory for up to 87.0%, thus proving this method to be effective and feasible.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 2377-2383
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.2377
• Online since: September 2013
• Authors: Peng Wu, Cheng Liu
• Keywords: Financial Distress, K-Means Clustering, Particle Swarm Optimization (PSO), Rough Set (RS)

References

[1] Ramser J, Foster L. (1931). A Demonstration of Ratio Analysis [R]. Urbana: University o f Illinois.

[2] Fitzpatrick P. A. (1931). Comparison of the Ratios of Successful Industrial Enterprises with Those of Failed Companies[R]. Washington: The Accountants Publishing Company.

[3] Merwin C. (1942). Financing Small Corporations: In Five Manufacturing Industries, 1926-1936 [M]. New York: National Bureau of Economic Research.

[4] Beaver W H. (1966). Financial ratios as predictors of failure, empirical research in accounting: Selected studies [J]. Journal of Accounting Research. pp.71-111.

DOI: 10.2307/2490171

[5] Altman E I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy [J]. The Journal of Finance, (23): 589-609.

DOI: 10.1111/j.1540-6261.1968.tb00843.x

[6] Ohlson J A. (1980). Financial ratios and probabilistic prediction of bankruptcy [J]. Journal of Accounting Research, 18: 109-131.

DOI: 10.2307/2490395

[7] Laitinen E, Laitinen T. (2000). Bankruptcy prediction: application of the Taylor's expansion in logistic regression [J]. International Review of Financial Analysis, 9: 327-349.

DOI: 10.1016/s1057-5219(00)00039-9

[8] Kidong Lee, David Booth, (2005). Pervaiz Alam. A comparison of supervised and unsupervised neural networks in predicting bankruptcy of Korean firms [J]. Expert Systems with applications. (29): 1-16.

DOI: 10.1016/j.eswa.2005.01.004

[9] I.M. Premachandra, Yao Chen, John Watson. (2011). DEA as a tool for predicting corporate failure and success: A case of bankruptcy assessment [J]. Omega. (39): 620-626.

DOI: 10.1016/j.omega.2011.01.002

[10] Krishma K, Murty M N. (1999). Genetic k-means algorithm [J]. IEEE Trans on System, Man, and Cybernetics. Part B, 29 (3): 433-439.

DOI: 10.1109/3477.764879