Soft Computing

, Volume 16, Issue 6, pp 919–942 | Cite as

Coevolution of lags and RBFNs for time series forecasting: L-Co-R algorithm

  • E. Parras-Gutierrez
  • M. Garcia-Arenas
  • V. M. RivasEmail author
  • M. J. del Jesus
Original Paper


This paper introduces Lags COevolving with Rbfns (L-Co-R), a coevolutionary method developed to face time-series forecasting problems. L-Co-R simultaneously evolves the model that provides the forecasted values and the set of time lags the model must use in the prediction process. Coevolution takes place by means of two populations that evolve at the same time, cooperating between them; the first population is composed of radial basis function neural networks; the second one contains the individuals representing the sets of lags. Thus, the final solution provided by the method comprises both the neural net and the set of lags that better approximate the time series. The method has been tested across 34 different time series datasets, and the results compared to 6 different methods referenced in literature, and with respect to 4 different error measures. The results show that L-Co-R outperforms the rest of methods, as the statistical analysis carried out indicates.


Neural networks Coevolutionary algorithms Time series forecasting Significant lags 



This work has been supported by the Caja Rural de Jaen and the University of Jaen (Spain) UJA-08-16-30 project, the regional project TIC-3928 (Feder Founds), and the Spanish project TIN 2008-06681-C06-02 (Feder Founds).


  1. Alcalá-Fdez J, Sánchez L, García S, del Jesus MJ, Ventura S, Garrell JM, Otero J, Romero C, Bacardit J, Rivas VM, Fernández JC, Herrera F (2009) Keel: a software tool to assess evolutionary algorithms for data mining problems. Soft Comput Fusion Found Methodol Appl 13:307–318. doi: 10.1007/s00500-008-0323-y Google Scholar
  2. Arizmendi CM, Sanchez J, Ramos NE, Ramos GI (1993) Time series predictions with neural nets: Application to airborne pollen forecasting. Int J Biometeorol 37:139–144. doi: 10.1007/BF01212623 CrossRefGoogle Scholar
  3. Armstrong JS, Collopy F (1992) Error measures for generalizing about forecasting methods: empirical comparisons. Int J Forecast 8:69–80CrossRefGoogle Scholar
  4. Assimakopoulos V, Nikolopoulos K (2000) The theta model: a decomposition approach to forecasting. Int J Forecast 16(4):521–530CrossRefGoogle Scholar
  5. Au CK, Leung HF (2007) Biasing mutations in cooperative coevolution. In: Proceedings of IEEE Congress on evolutionary computation, CEC 2007, pp 828–835Google Scholar
  6. Bezerianos A, Papadimitriou S, Alexopoulos D (1999) Radial basis function neural networks for the characterization of heart rate variability dynamics. Artif Intell Med 15(3):215–234. doi: 10.1016/S0933-3657(98)00055-4 CrossRefGoogle Scholar
  7. Bowerman BL, O’Connell RT, Koehler AB (2004) Forecasting: methods and applications. Thomson Brooks/Cole: Belmont, CAGoogle Scholar
  8. Box GEP, Jenkins GM (1976) Time series analysis: forecasting and control. Holden Day, San FranciscozbMATHGoogle Scholar
  9. Bradley MD, Jansen DW (2004) Forecasting with a nonlinear dynamic model of stock returns and industrial production. Int J Forecast 20(2):321–342. doi: 10.1016/j.ijforecast.2003.09.007 CrossRefGoogle Scholar
  10. Brockwell P, Hyndman R (1992) On continuous-time threshold autoregression. Int J Forecast 8(2):157–173. doi: 10.1016/0169-2070(92)90116-Q CrossRefGoogle Scholar
  11. Broomhead D, Lowe D (1988) Multivariable functional interpolation and adaptive networks. Complex Syst 2:321–355MathSciNetzbMATHGoogle Scholar
  12. Brown R (1959) Statistical forecasting for inventory control. McGraw-Hill, New YorkGoogle Scholar
  13. Carse B, Fogarty T (1996) Fast evolutionary learning of minimal radial basis function neural networks using a genetic algorithm. In: Proceedings of evolutionary computing. Lecture dois in Computer Science, vol 1143, pp 1–22. Springer, Berlin. doi: 10.1007/BFb0032769
  14. Castillo PA, Merelo JJ, Prieto A, Rivas VM, Romero G (2000) G-prop: global optimization of multilayer perceptrons using gas. Neurocomputing 35:149–163. doi: 10.1016/S0925-2312(00)00302-7 zbMATHCrossRefGoogle Scholar
  15. Chan KS, Tong H (1986) On estimnating thresholds in autoregressive models. J Time Ser Anal 7(3):179–190. doi: 10.1111/j.1467-9892.1986.tb00501.x MathSciNetzbMATHCrossRefGoogle Scholar
  16. Chatterjee A, Siarry P (2006) Nonlinear inertia weight variation for dynamic adaptation in particle swarm optimization. Comput Oper Res 33:859–871. doi: 10.1016/j.cor.2004.08.012 zbMATHCrossRefGoogle Scholar
  17. Clements MP, Franses PH, Swanson NR (2004) Forecasting economic and financial time-series with non-linear models. Int J Forecast 20(2):169–183. doi: 10.1016/j.ijforecast.2003.10.004 CrossRefGoogle Scholar
  18. Crone S, Hibon M, Nikolopoulos K (2011) Advances in forecasting with neural networks? Empirical evidence from the nn3 competition on time series prediction. Int J Forecast 27(3):635–660Google Scholar
  19. Dash PK, Liew AC, Rahman S, Ramakrishna G (1995) Building a fuzzy expert system for electric load forecasting using a hybrid neural network. Exp Syst Appl 9(3):407–421. doi: 10.1016/0957-4174(95)00013-Y CrossRefGoogle Scholar
  20. Dawson CW, Wilby RL, Harpham C, Brown MR, Cranston E, Darby EJ (2000) Modelling ranunculus presence in the rivers test and itchen using artificial neural networks. In: Proceedings of international conference on geocomputationGoogle Scholar
  21. de A Araújo R (2010a) Hybrid intelligent methodology to design translation invariant morphological operators for brazilian stock market prediction. Neural Netw 23:1238–1251CrossRefGoogle Scholar
  22. de A Araújo R (2010b) A quantum-inspired evolutionary hybrid intelligent approach for stock market prediction. Int J Intell Comput Cybern 3(10):24–54MathSciNetzbMATHCrossRefGoogle Scholar
  23. de A Araújo R (2010c) Swarm-based translation-invariant morphological prediction method for financial time series forecasting. Inform Sci 180:4784–4805MathSciNetzbMATHCrossRefGoogle Scholar
  24. de A Araújo R (2011) Translation invariant morphological time-lag added evolutionary forecasting method for stock market prediction. Exp Syst Appl 38:2835–2848. doi: 10.1016/j.eswa.2010.08.076 CrossRefGoogle Scholar
  25. Derrac J, García S, Herrera F (2010) Ifs-coco: instance and feature selection based on cooperative coevolution with nearest neighbor rule. Pattern Recogn 43(6):2082–2105. doi: 10.1016/j.patcog.2009.12.012 zbMATHCrossRefGoogle Scholar
  26. Du H, Zhang N (2008) Time series prediction using evolving radial basis function networks with new encoding scheme. Neurocomputing 71:1388–1400. doi: 10.1016/j.neucom.2007.06.004 CrossRefGoogle Scholar
  27. Eshelman LJ (1991) The chc adptive search algorithm: how to have safe search when engaging in nontraditional genetic recombination. In: Proceedings of first workshop on foundations of genetic algorithms, Morgan Kaufmann, Menlo Park, pp 265–283Google Scholar
  28. Ferreira T, Vasconcelos G, Adeodato P (2008) A new intelligent system methodology for time series forecasting with artificial neural networks. Neural Process Lett 28(2):113–129CrossRefGoogle Scholar
  29. Fildes R (1983) An evaluation of bayesian forecasting. J Forecast 2(2):137–150. doi: 10.1002/for.3980020205 CrossRefGoogle Scholar
  30. Fildes R (1992) The evaluation of extrapolative forecasting methods. Int J Forecast 8(1):81–98. doi: 10.1016/0169-2070(92)90009-X CrossRefGoogle Scholar
  31. Fildes R, Nikolopoulos K, Crone SF, Syntetos AA (2008) Forecasting and operational research: a review. J Oper Res Soc 59:1150–1172zbMATHCrossRefGoogle Scholar
  32. Fu X, Wang L (2003) Data dimensionality reduction with application to simplifying rbf network structure and improving classification performance. IEEE Trans Syst Man Cybern Part B 33:399–409. doi: 10.1109/TSMCB.2003.810911 CrossRefGoogle Scholar
  33. García S, Fernández A, Luengo J, Herrera F (2009) A study of statistical techniques and performance measures for genetics-based machine learning: accuracy and interpretability. Soft Comput 13:959–977. doi: 10.1007/s00500-008-0392-y CrossRefGoogle Scholar
  34. García Pajares R, Benitez J, Sainz Palmero G (2008) Feature selection form time series forecasting: a case study. In: Eighth international conference on hybrid intelligent systems, pp 555–560Google Scholar
  35. Garcia-Pedrajas N, Hervas-Martinez C, Ortiz-Boyer D (2005) Cooperative coevolution of artificial neural network ensembles for pattern classification. IEEE Trans Evol Comput 9:271–302. doi: 10.1109/TEVC.2005.844158 CrossRefGoogle Scholar
  36. García-Pedrajas N, del Castillo JR, Ortiz-Boyer D (2010) A cooperative coevolutionary algorithm for instance selection for instance-based learning. Mach Learn 78:381–420. doi: 10.1007/s10994-009-5161-3 CrossRefGoogle Scholar
  37. Gardner ES (1985) Exponential smoothing: the state of the art. J Forecast 4(1):1–28. doi: 10.1002/for.3980040103 CrossRefGoogle Scholar
  38. Gooijer JGD, Hyndman RJ (2006) 25 years of time series forecasting. Int J Forecast 22(3):443–473CrossRefGoogle Scholar
  39. Granger C, Tersvirta T (1993) Modelling non-linear economic relationships. Oxford University Press, OxfordGoogle Scholar
  40. Harpham C, Dawson CW (2006) The effect of different basis functions on a radial basis function network for time series prediction: a comparative study. Neurocomputing 69:2161–2170. doi: 10.1016/j.neucom.2005.07.010 CrossRefGoogle Scholar
  41. Harpham C, Dawson CW, Brown MR (2004) A review of genetic algorithms applied to training radial basis function networks. Neural Comput Appl 13:193–201. doi: 10.1007/s00521-004-0404-5 CrossRefGoogle Scholar
  42. Harrison PJ, Stevens CF (1976) Bayesian forecasting. J Royal Stat Soc Ser B (Methodological) 38(3):205–247MathSciNetzbMATHGoogle Scholar
  43. Harvey AC (1984) A unified view of statistical forecasting procedures. J Forecast 3(3):245–275. doi: 10.1002/for.3980030302 MathSciNetCrossRefGoogle Scholar
  44. Hippert HS, Taylor JW (2010) An evaluation of bayesian techniques for controlling model complexity and selecting inputs in a neural network for short-term load forecasting. Neural Netw 23(3):386–395. doi: 10.1016/j.neunet.2009.11.016 CrossRefGoogle Scholar
  45. Holland JH (1975) Adaptation in natural and artificial systems. The University of Michigan Press, Ann ArborGoogle Scholar
  46. Holm S (1979) A simple sequentially rejective multiple test procedure. Scand J Stat 6:65–70MathSciNetzbMATHGoogle Scholar
  47. Hyndman RJ, Billah B (2003) Unmasking the theta method. Int J Forecast 19(2):287–290zbMATHCrossRefGoogle Scholar
  48. Hyndman RJ, Koehler AB (2006) Another look at measures of forecast accuracy. Int J Forecast 22(4):679–688. doi: 10.1016/j.ijforecast.2006.03.001 CrossRefGoogle Scholar
  49. Jain A, Kumar AM (2007) Hybrid neural network models for hydrologic time series forecasting. Appl Soft Comput 7(2):585–592. doi: 10.1016/j.asoc.2006.03.002 MathSciNetCrossRefGoogle Scholar
  50. Kavaklioglu K (2011) Modeling and prediction of turkey’s electricity consumption using support vector regression. Appl Energ 88(1):368–375. doi: 10.1016/j.apenergy.2010.07.021 CrossRefGoogle Scholar
  51. Lee CM, Ko CN (2009) Time series prediction using rbf neural networks with a nonlinear time-varying evolution pso algorithm. Neurocomputing 73(1–3):449–460. doi: 10.1016/j.neucom.2009.07.005 CrossRefGoogle Scholar
  52. Li M, Tian J, Chen F (2008) Improving multiclass pattern recognition with a co-evolutionary rbfnn. Pattern Recogn Lett 29(4):392–406. doi: 10.1016/j.patrec.2007.10.019 CrossRefGoogle Scholar
  53. Lukoseviciute K, Ragulskis M (2010) Evolutionary algorithms for the selection of time lags for time series forecasting by fuzzy inference systems. Neurocomputing 73:2077–2088CrossRefGoogle Scholar
  54. Ma X, Wu HX (2010) Power system short-term load forecasting based on cooperative co-evolutionary immune network model. In: Proceedings of 2nd international conference on education technology and computer (ICETC), pp 582–585Google Scholar
  55. Makridakis SG, Hibon M (2000) The m3-competition: results, conclusions and implications. Int J Forecast 16(4):451–476CrossRefGoogle Scholar
  56. Makridakis SG, Andersen A, Carbone R, Fildes R, Hibon M, Lewandowski R, Newton J, Parzen E, Winkler R (1982) The accuracy of extrapolation (time series) methods: results of a forecasting competition. J Forecast 1(2):111–153. doi: 10.1002/for.3980010202 CrossRefGoogle Scholar
  57. Martínez-Estudillo A, Martínez-Estudillo F, Hervás-Martínez C, García-Pedrajas N (2006) Evolutionary product unit based neural networks for regression. Neural Netw 19(4):477–486. doi: 10.1016/j.neunet.2005.11.001 zbMATHCrossRefGoogle Scholar
  58. Maus A, Sprott JC (2011) Neural network method for determining embedding dimension of a time series. Commun Nonlinear Sci Numer Simul 16(8):3294–3302MathSciNetzbMATHCrossRefGoogle Scholar
  59. Merelo JJ, Prieto A (1995) G-lvq, a combination of genetic algorithms and lvq. In: Proceedings of artificial neural nets and genetic algorithms, Springer, Berlin, pp 92–95Google Scholar
  60. Panait L, Wiegand RP, Luke S (2003) Improving coevolutionary search for optimal multiagent behaviors. In: Proceedings of the eighteenth international joint conference on artificial intelligence, Morgan Kaufmann, Menlo Park, pp 653–658Google Scholar
  61. Paredis J (1995) Coevolutionary computation. Artif Life 2(4):355–375. doi: 10.1162/artl.1995.2.4.355 CrossRefGoogle Scholar
  62. Pena D (2005) Análisis de Series Temporales. Alianza EditorialGoogle Scholar
  63. Perez-Godoy MD, Aguilera JJ, Berlanga FJ, Rivas VM, Rivera AJ (2008) A preliminary study of the effect of feature selection in evolutionary rbfn design. In: Proceedings of information processing and management of uncertainty in knowledge-based system, pp 1151–1158Google Scholar
  64. Perez-Godoy MD, Pérez-Recuerda P, Frías M, Rivera AJ, Carmona C, Parras M (2010a) Co2rbfn for short and medium term forecasting of the extra-virgin olive oil price. In: González J, Pelta D, Cruz C, Terrazas G, Krasnogor N (eds) Proceedings of nature inspired cooperative strategies for optimization (NICSO 2010), Studies in Computational Intelligence, vol 284, pp 113–125. Springer, Berlin. doi: 10.1007/978-3-642-12538-6_10
  65. Perez-Godoy MD, Rivera A, Berlanga FJ, del Jesus MJ (2010b) Co2rbfn: an evolutionary cooperative–competitive rbfn design algorithm for classification problems. Soft Comput Fusion Found Methodol Appl 14:953–971. doi: 10.1007/s00500-009-0488-z Google Scholar
  66. Potter M, Jong KD (1994) A cooperative coevolutionary approach to function optimization. In: Davidor Y, Schwefel HP, Mnner R (eds) Proceedings of parallel problem solving from nature PPSN III, Lecture Notes in Computer Science, vol 866, pp 249–257. Springer, Berlin. doi: 10.1007/3-540-58484-6_269
  67. Potter M, Jong KD (2000) Cooperative coevolution: an architecture for evolving coadapted subcomponents. Evol Comput 8(1):1–29. doi: 10.1162/106365600568086 CrossRefGoogle Scholar
  68. Qian-Li M, Qi-Lun Z, Hong P, Tan-Wei Z, Jiang-Wei Q (2008) Multi-step-prediction of chaotic time series based on co-evolutionary recurrent neural network. Chin Phys B 17(2). doi: 10.1088/1674-1056/17/2/031
  69. Qiu W, Liu X, Li H (2011) A generalized method for forecasting based on fuzzy time series. Exp Syst Appl 38(8):10446–10453. doi: 10.1016/j.eswa.2011.02.096 CrossRefGoogle Scholar
  70. Rivas VM, Merelo JJ, Castillo PA, Arenas MG, Castellano JG (2004) Evolving rbf neural networks for time-series forecasting with evrbf. Inform Sci 165(3–4):207–220. doi: 10.1016/j.ins.2003.09.025 MathSciNetCrossRefGoogle Scholar
  71. Rivas VM, Arenas MG, Merelo JJ, Prieto A (2007) Evrbf: evolving rbf neural networks for classification problems. In: Proceedings of the 7th conference on 7th WSEAS international conference on applied informatics and communications, Stevens Point, Wisconsin, USA, vol 7, pp 98–103Google Scholar
  72. Rivera AJ, Rojas I, Ortega J, del Jesus MJ (2007) A new hybrid methodology for cooperative-coevolutionary optimization of radial basis function networks. Soft Comput Fusion Found Methodol Appl 11:655–668. doi: 10.1007/s00500-006-0128-9 Google Scholar
  73. Rustagi JS (1994) Optimization techniques in statistics. Academic Press, BostonzbMATHGoogle Scholar
  74. Samanta B (2011) Prediction of chaotic time series using computational intelligence. Exp Syst Appl 38(9):11406–11411. doi: 10.1016/j.eswa.2011.03.013 CrossRefGoogle Scholar
  75. Sarantis N (2001) Nonlinearities, cyclical behaviour and predictability in stock markets: international evidence. Int J Forecast 17(3):459–482. doi: 10.1016/S0169-2070(01)00093-0 CrossRefGoogle Scholar
  76. Sergeev S, Mahotilo K, Voronovsky G, Petrashev S (1998) Genetic algorithm for training dynamical object emulator based on rbf neural network. Int J Appl Electromagn Mech 9:65–74Google Scholar
  77. Sheskin D (2006) Handbook of parametric and nonparametric statistical procedures. Chapman & Hall/CRC, LondonGoogle Scholar
  78. Sheta AF, Jong KD (2001) Time-series forecasting using ga-tuned radial basis functions. Inform Sci 133(3-4):221–228. doi: 10.1016/S0020-0255(01)00086-X zbMATHCrossRefGoogle Scholar
  79. Snyder RD (1985) Recursive estimation of dynamic linear models. J Royal Stat Soc Ser B (Methodological) 47:272–276. Google Scholar
  80. Sun ZL, Huang D, Zheng CH, Shang L (2006) Optimal selection of time lags for tdsep based on genetic algorithm. Neurocomputing 69(7–9):884–887CrossRefGoogle Scholar
  81. Takens F (1980) Detecting strange attractor in turbulence. In: Dynamical systems and turbulence. Lecture notes in mathematics, vol 898. Springer, New York, NY, pp 366–381Google Scholar
  82. Tan KC, Yang YJ, Goh CK (2006) A distributed cooperative co-evolutionary algorithm for multi-objective optimization. IEEE Trans Evol Comput 10:527–549. doi: 10.1109/TEVC.2005.860762 CrossRefGoogle Scholar
  83. Tanaka N, Okamoto H, Naito M (2001) Estimating the active dimension of the dynamics in a time series based on a information criterion. Phys D 158:19–31zbMATHCrossRefGoogle Scholar
  84. Tang Z, de Almeida C, Fishwick PA (1991) Time series forecasting using neural networks vs. boxjenkins methodology. Simulation 57:303–310CrossRefGoogle Scholar
  85. Tong H (1978) On a threshold model. Pattern Recogn Signal Process NATO ASI Ser E Appl Sci 29:575–586Google Scholar
  86. Tong H (1983) Threshold models in non-linear time series analysis. In: Lecture notes in statistics, vol 21. Springer, BerlinGoogle Scholar
  87. Valenzuela O, Rojas I, Rojas F, Pomares H, Herrera LJ, Guillen A, Marquez ML, Pasadas M (2008) Hybridization of intelligent techniques and arima models for time series prediction. Fuzzy Sets Syst 159(7):821–845. doi: 10.1016/j.fss.2007.11.003 MathSciNetzbMATHCrossRefGoogle Scholar
  88. Wang CC (2011) A comparison study between fuzzy time series model and arima model for forecasting taiwan export. Exp Syst Appl 38(8):9296–9304CrossRefGoogle Scholar
  89. Wang LX, Mendel JM (2002) Generating fuzzy rules by learning from examples. IEEE Trans Syst Man Cybern 22:1414–1427. doi: 10.1109/21.199466 MathSciNetCrossRefGoogle Scholar
  90. Whitehead BA, Choate TD (1996) Cooperative-competitive genetic evolution of radial basis function centers and widths for time series prediction. IEEE Trans Neural Netw 7:869–880. doi: 10.1109/72.508930 CrossRefGoogle Scholar
  91. Wichern DW, Jones RH (1977) Assessing the impact of market disturbances using intervention analysis. Manag Sci 24:329–337zbMATHCrossRefGoogle Scholar
  92. Wiegand RP, Liles WC, De Jong K (2001) An empirical analysis of collaboration methods in cooperative coevolutionary algorithms. In: Proceedings of the genetic and evolutionary computation conference, Morgan Kaufmann, Menlo Park, pp 1235–1242Google Scholar
  93. Winters PR (1960) Forecasting sales by exponentially weighted moving averages. Manag Sci 6:324–342. MathSciNetzbMATHCrossRefGoogle Scholar
  94. Xue Y, Watton J (1998) Dynamics modelling of fluid power systems applying a global error descent algorithm to a self-organising radial basis function network. Mechatronics 8(7):727–745. doi: 10.1016/S0957-4158(98)00024-5 CrossRefGoogle Scholar
  95. Yu THK, Huarng KH (2010) A neural network-based fuzzy time series model to improve forecasting. Exp Syst Appl 37(4):3366–3372. doi: 10.1016/j.eswa.2009.10.013 CrossRefGoogle Scholar
  96. Zar J (1999) Biostatistical analysis. Prentice Hall, Englewood CliffsGoogle Scholar
  97. Zhang G, Hu MY (1998) Neural network forecasting of the british pound/us dollar exchange rate. Omega Int J Manag Sci 26(4):495–506. doi: 10.1016/S0305-0483(98)00003-6 CrossRefGoogle Scholar
  98. Zhang G, Patuwo BE, Hu MY (1998) Forecasting with artificial neural networks: the state of the art. Int J Forecast 14(1):35–62. doi: 10.1016/S0169-2070(97)00044-7 CrossRefGoogle Scholar
  99. Zhang GP, Qi M (2005) Neural network forecasting for seasonal and trend time series. Eur J Oper Res 160(2):501–514. doi: 10.1016/j.ejor.2003.08.037 zbMATHCrossRefGoogle Scholar
  100. Zhu S, Wang J, Zhao W, Wang J (2011) A seasonal hybrid procedure for electricity demand forecasting in china. Appl Energ 88(11):3807–3815. doi: 10.1016/j.apenergy.2011.05.005 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • E. Parras-Gutierrez
    • 1
  • M. Garcia-Arenas
    • 2
  • V. M. Rivas
    • 1
    Email author
  • M. J. del Jesus
    • 1
  1. 1.Department of Computer SciencesJaenSpain
  2. 2.Department of Computers, Architecture and TechnologyGranadaSpain

Personalised recommendations