Fuzzy neural and chaotic searching hybrid algorithm and its application in electric customers’s credit risk evaluation

Abstract

To evaluate the credit risk of customers in power market precisely, the new chaotic searching and fuzzy neural network (FNN) hybrid algorithm were proposed. By combining with the chaotic searching, the learning ability of the FNN was markedly enhanced. Customers’ actual credit flaw data of power supply enterprises were collected to carry on the real evaluation, which can be treated as example for the model. The result shows that the proposed method surpasses the traditional statistical models in regard to the precision of forecasting and has a practical value. Compared with the results of ordinary FNN and ANN, the precision of the proposed algorithm can be enhanced by 2.2% and 4.5%, respectively.

Keywords

power supply enterprise / credit-risk / fuzzy neural network / chaotic searching

References

[1] JarrowR., LandoD., TurnbullS.. A Markov model of the term structure of credit risk spreads[J]. Review of Financial Studies, 1997, 10(2): 481-523

[2] BaesensB., van GestelT., ViaeneS., et al.. Benchmarking state of the art classification algorithms for credit scoring[J]. Journal of Operations Research Society, 2003, 54(6): 627-635

[3] AtiyaA. F.. Bankruptcy prediction for credit risk using neural networks: A survey and new results[J]. IEEE Transactions on Neural Networks, 2001, 12(4): 929-935

[4] RochaK., FranciscoA., GarciaA.. Credit risk in the pool—implications for private capital investments in Brazilian power generation[J]. Energy policy, 2006, 34(18): 3827-3835

[5] DuffieD.. Credit risk modeling with affine processes[J]. Journal of Banking and Finance, 2005, 29(11): 2751-2802

[6] WangL., ZhengD. Z., LinQ. S.. Survey on chaotic optimization methods[J]. Comput Technol Automat, 2001, 20(1): 1-5

[7] TangW., LiD.-pu.. Chaotic optimization for economic dispatch of power systems[J]. Proceedings of CSEE, 2000, 20(10): 36-40

[8] MaoY.-l., ZhangG.-z., ZhuB., et al.. Economic load dispatch of power systems based on chaotic simulated annealing neural network model[J]. Proceedings of CSEE, 2005, 25(3): 65-70

[9] JuangC. F.. A TSK-type recurrent fuzzy network for dynamic systems processing by neural network and genetic algorithms[J]. Fuzzy Syst, 2002, 10(2): 155-170