Chinese credit scale prediction using M-estimator based robust radial basis function neural networks

ZHANG Ying, HONG Zhen-yu, JIANG Wei

Systems Engineering - Theory & Practice ›› 2014, Vol. 34 ›› Issue (12) : 3022-3033.

PDF(985 KB)
PDF(985 KB)
Systems Engineering - Theory & Practice ›› 2014, Vol. 34 ›› Issue (12) : 3022-3033. DOI: 10.12011/1000-6788(2014)12-3022

Chinese credit scale prediction using M-estimator based robust radial basis function neural networks

  • ZHANG Ying1,2, HONG Zhen-yu2, JIANG Wei2
Author information +
History +

Abstract

An M-estimator based robust radial basis function (RBF) learning algorithm with growing and pruning techniques using the concept of statistical contribution was proposed. This method not only eliminates the influence of noise and outliers, but also dynamically adjusts the number of neurons to approach an appropriate size of the network in estimating the parameters at the same time, and so improves the speeds of learning and convergence. Compared with SARIMA and SVR approaches based on empirical RMB monthly loan data series in China, the proposed method has the strongest forecast ability among all methods, and has an important value in applications to improve the effectiveness and perspectiveness of the monetary policy.

Key words

statistical contribution / M-estimator / radial basis function (RBF) / encompassing test

Cite this article

Download Citations
ZHANG Ying , HONG Zhen-yu , JIANG Wei. Chinese credit scale prediction using M-estimator based robust radial basis function neural networks. Systems Engineering - Theory & Practice, 2014, 34(12): 3022-3033 https://doi.org/10.12011/1000-6788(2014)12-3022

References

[1] Clements M P, Fred J, Herman O S. An evaluation of the forecasts of the federal reserve: A pooled approach[J]. Journal of Applied Econometrics, 2007, 22(1): 121-136.
[2] Edge R M, Michael T K, Jean P L. A comparison of forecast performance between Federal Reserve staff forecasts, simple reduced-form models, and a DSGE model[J]. Journal of Applied Econometrics, 2010, 25(2): 720-754.
[3] Cassino V, Misich P, Barry J. Forecasting the demand of currency[R]. Reserve Bank Bulletin, Reserve Bank of New Zealand, 1997, 60: 27-33.
[4] Chatfield C. Time-series forecasting[M]. Chapman and Hall/CRC Press, 2000.
[5] Kuan C M, Liu T. Forecasting exchange rates using feedforward and recurrent neural networks[J]. Journal of Applied Econometrics, 1995, 10(4): 347-364.
[6] Park J, Sandberg I W. Universal approximation using radial-basis-function networks[J]. Neural Computation, 1991, 3: 246-257.
[7] Kim K. Financial time series forecasting using support vector machines[J]. Neurocomputing, 2003, 55: 307-319.
[8] 陈诗一.汇率预测:一个新的非参数支持向量回归方法[J].数量经济技术经济研究, 2007(5): 23-31. Chen Shiyi. Forecasting exchange rates: A new nonparametric support vector regression rate[J]. Quantitative & Technical Economics, 2007(5): 23-31.
[9] Bors A G, Pitas I. Median radial basis function neural network[J]. IEEE Transactions on Neural Networks, 1996, 7(6): 1351-1364.
[10] Yin H, Allinson N M. Self-organizing mixture networks for probability density estimation[J]. IEEE Transactions on Neural Networks, 2001, 12(2): 405-411.
[11] Uykan Z C, Guzelis M, Koivo H N. Analysis of input-output clustering for determining centers of RBFN[J]. IEEE Transactions on Neural Networks, 2000, 11(4): 851-858.
[12] Gonzalez J, Rojas I, Ortega J, et al. Multi-objective evolutionary optimization of the size, shape, and position parameters of radial basis function networks for function approximation[J]. IEEE Transactions on Neural Networks, 2003, 14(6): 1478-1495.
[13] Leung F H F, Lam H K, Ling S H, et al. Tuning of the structure and parameters of a neural network using an improved genetic algorithm[J]. IEEE Transactions on Neural Networks, 2003, 14(1): 79-88.
[14] Platt J. A resource-allocating network for function interpolation[J]. Neural Computation, 1991, 3: 213-225.
[15] Esposito A, Marinaro M, Oricchio D, et al. Approximation of continuous and discontinuous mappings by a growing neural RBF-based algorithm[J]. Neural Networks, 2000, 13(6): 651-665.
[16] Mozer M C, Smolensky P. Skeletonization: A technique for trimming the fat from a network via relevance assessment[C]// Advances in Neural Information Processing Systems, 1988: 107-115.
[17] Fu X J, Wang L P. Data dimensionality reduction with application to simplifying RBF network structure and improving classification performance[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, 2003, 33(3): 399-409.
[18] Feng H M. Self-generation RBFNs using evolutional PSO learning[J]. Neurocomputing, 2006, 70(1-3): 241-251.
[19] Balestrassi P P, Popova E, Paiva A P, et al. Design of experiments on neural network's training for nonlinear time series forecasting[J]. Neurocomputing, 2009, 72(4): 1160-1178.
[20] Murata N S, Yoshizawa S A. Network information criterion-determining the number of hidden units for an artificial neural network model[J]. IEEE Transactions on Neural Networks, 1994, 5: 865-872.
[21] 甘敏, 彭辉, 陈晓红. RBF-AR模型在非线性时间序列预测中的应用[J].系统工程理论与实践, 2010, 30(6): 1055-1061. Gan Min, Peng Hui, Chen Xiaohong. RBF-AR model-based nonlinear time series prediction[J]. Systems Engineering - Theory & Practice, 2010, 30(6): 1055-1061.
[22] 魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社, 2005: 33-35. Wei Haikun. The structural design of neural network theory and method[M]. Beijing: Defense Industry Publishing House, 2005: 33-35.
[23] Huber P J. Robust statistics[M]. New York: John Wiley and Sons, 1984.
[24] Chng E S, Chen S, Mulgrew B. Gradient radial basis function networks for nonlinear and nonstationary time series prediction[J]. IEEE Transactions on Neural Networks, 1996, 7(1): 190-194.
[25] Leung H, Wang L T. Prediction of noisy chaotic time series using an optimal radial basis functions neural network[J]. IEEE Transactions on Neural Networks, 2001, 12(5): 1163-1172.
[26] Chen D S, Jain R C. A robust back-propagation learning algorithm for function approximation[J]. IEEE Transactions on Neural Networks, 1994, 5: 467-479.
[27] David V, Sanchez A. Robustization of learning method for RBF networks[J]. Neurocomputing, 1995, 9: 85-94.
[28] Amidan B G, Thomas A F, Scott K C. Data outlier detection using the Chebyshev theorem[J]. 2005 IEEE Conference on Aerospace, 2005: 3814-3819.
[29] Lu Y W, Sundararajan N, Saratchandran P. A sequential learning scheme for function approximation using minimal radial basis function (RBF) neural networks[J]. Neural Computing, 1997, 9: 461-478.
[30] Gonzalo J, Pitarakis J Y. Lag length estimation in large dimensional systems[J]. Journal of Time Series Analysis, 2002, 23(4): 401-423.
[31] Zhang G P, Patuwo B E, Hu Y. A simulation study of artificial neural networks for nonlinear time-series forecasting[J]. Computers & Operations Research, 2001, 28(5): 381-396.
[32] Walczak S. An empirical analysis of data requirements for financial forecasting with neural networks[J]. Journal of Management Information Systems, 2001, 17(4): 203-222.
[33] Huang W, Nakamori Y, Wang S. A general approach based on autocorrelation to determine input variables of neural networks for time series forecasting[J]. Journal of Systems Science and Complexity, 2004, 17(3): 297-305.
[34] Tsay R S. Outliers, level shifts, and variance changes in time series[J]. Journal of Forecasting, 1988, 7: 1-20.
[35] Hodge V J, Austin J. A survey of outlier detection methodologies[J]. Artificial Intelligence Review, 2004, 22: 85-126.
[36] Khamis A, Ismail Z, Haron K, et al. The effects of outliers data on neural network performance[J]. Journal of Applied Science, 2005, 5(8): 1394-1398.
[37] Stone M. Asymptotic for and against cross-validation[J]. Biometrika, 1977, 64(1): 29-35.
[38] Masters T. Practical neural network recipes in C++[M]. New York: Academic Press, 1993.
[39] Box G J, Reinsel G. Time series analysis: Forecasting & control[M]. Prentice Hall, 1994.
[40] Harvey D I, Stephen J L, Paul N. Tests for forecast encompassing[J]. Journal of Business and Economic Statistics, 1998, 16(2): 254-259.
[41] 柯莱蒙兹,韩德勤,陆懋祖.预测经济时间序列[M]. 北京:北京大学出版社, 2008.Clements M P, Hendry D F, Lu Maozu. Forecasting economic time series[M]. Beijing: Peking University Publishing House, 2008.
[42] Wang S Y. TEI@I: A new methodology for studying complex systems[R]. The International Workshop on Complexity Science, Tsukuba, Japan, 2004.
[43] 汪寿阳, 余乐安, 黎建强. TEI@I方法论及其在外汇汇率预测中的应用[J].管理学报, 2007(1): 21-27. Wang Shouyang, Yu Lean, Li Jianqiang. TEI@I methodology and its application to exchange rates prediction[J]. Chinese Journal of Management, 2007(1): 21-27.
PDF(985 KB)

327

Accesses

0

Citation

Detail

Sections
Recommended

/