Document Type: Research Paper


1 Industrial Engineering Department, Isfahan University of Technol- ogy, Isfahan, Iran

2 Industrial Engineering Department, Isfahan University of Technology, Isfahan, Iran

3 Industrial Engineering Department, Isfahan University of Tech- nology, Isfahan, Iran


Improving time series forecasting
accuracy is an important yet often difficult task.
Both theoretical and empirical findings have
indicated that integration of several models is an effective
way to improve predictive performance, especially
when the models in combination are quite different. In this paper,
a model of the hybrid artificial neural networks and
fuzzy model is proposed for time series forecasting, using
autoregressive integrated moving average models. In the proposed
model, by first modeling the linear components, autoregressive integrated moving average models are
combined with the these hybrid models to yield a
more general and accurate forecasting model than the
traditional hybrid artificial neural networks and fuzzy models. Empirical results for  financial
time series forecasting indicate that the proposed model exhibits
effectively improved forecasting accuracy and hence is an appropriate forecasting tool for financial time
series forecasting.


bibitem{1} A. R. Arabpour and M. Tata, {it Estimating the parameters of a
fuzzy linear regression model}, Iranian Journal of Fuzzy Systems,
{bf5} (2008), 1-19.
bibitem{2} G. Armano, M. Marchesi and A. Murru, {it A
hybrid genetic-neural architecture for stock indexes forecasting},
Information Sciences, {bf170} (2005), 3-33.
bibitem{1} J. M. Bates and W. J.
Granger, {it The combination of forecasts}, Operation Research,
{bf20} (1969), 451-468.
bibitem{1} P. Box and G. M. Jenkins, {it Time
series analysis: forecasting and control}, Holden-day Inc, San
Francisco, CA, 1976.
bibitem{1} M. C. Brace, J. Schmidt and M. Hadlin, {it
Comparison of the forecasting accuracy of neural networks with
other established techniques}, In: Proceedings of the First Forum
on Application for weight elimination, IEEE Transactions on Neural
Networks of Neural Networks to Power Systems, Seattle, WA (1991),
bibitem{1} P. Chang, C. Liu and Y. Wang, {it A hybrid model by
clustering and evolving fuzzy rules for sales decision supports in
printed circuit board industry}, Decision Support Systems, {bf42}
(2006), 1254-1269.
bibitem{1} S. M. Chen, {it Forecasting enrollments
based on fuzzy time series}, Fuzzy Sets and Systems, {bf81(3)}
(1996), 311--319, 1996.
bibitem{1} K. Y. Chen and C. H. Wang, {it A hybrid
SARIMA and support vector machines in forecasting the production
values of the machinery industry in Taiwan}, Expert Systems with
Applications, {bf32} (2007), 254-264.
bibitem{1} S. M. Chen and N. Y. Chung,
{it Forecasting enrollments using high-order fuzzy time series
and genetic algorithms}, International J. Intell. Syst., {bf21}
(2006), 485-501.
bibitem{1} R. Clemen, {it Combining forecasts: a
review and annotated bibliography with discussion}, International
Journal of Forecasting, {bf5} (1989), 559-608.
bibitem{1} J. W. Denton,
{it How good are neural networks for causal forecasting?}, The
Journal of Business Forecasting, {bf14(2)} (1995), 17-20.
P. A. Fishwick, {it Neural network models in simulation: a
comparison with traditional modeling approaches}, In: Proceedings
of Winter Simulation Conference, Washington, D. C., (1989),
bibitem{1} W. R. Foster, F. Collopy and L. H. Ungar, {it
Neural network forecasting of short, noisy time series}, Computers
and Chemical Engineering, {bf16(4)} (1992), 293-297.
I. Ginzburg and D. Horn, {it Combined neural networks for time
series analysis}, Neural Information Processing Systems, {bf6}
(1994), 224-231.
bibitem{1} T. H. Hann and E. Steurer, {it Much
ado about nothing? exchange rate forecasting: neural networks vs.
linear models using monthly and weekly data}, Neurocomputing,
{bf10} (1996), 323-339.
bibitem{1} M. Haseyama and H. Kitajima,
{it An ARMA order selection method with fuzzy reasoning}, Signal
Process, {bf81} (2001), 1331-1335.
bibitem{1} H. Hassanpour, H.
R. Maleki and M. A. Yaghoobi, {it A note on evaluation of fuzzy
linear regression models by comparing membership functions},
Iranian Journal of Fuzzy Systems, {bf6} (2009), 1-6.
M. Hibon and T. Evgeniou, {it To combine or not to combine:
selecting among forecasts and their combinations}, International
Journal of Forecasting, {bf21} (2005), 15-24.
bibitem{1} C. M.
Hurvich and C. L. Tsai, {it Regression and time series model
selection in small samples}, Biometrica, {bf76(2)} (1989),
bibitem{1} H. B. Hwang, {it Insights into
neural-network forecasting time series corresponding to ARMA(p; q)
structures}, Omega, {bf29} (2001), 273-289.
bibitem{1} H.
Ishibuchi and H. Tanaka, {it Interval regression analysis based on
mixed 0-1 integer programming problem}, J. Japan Soc. Ind. Eng,
{bf40(5)} (1988), 312-319.
bibitem{1} J. S. R. Jang, {it ANFIS:
adaptive-network-based fuzzy inference system}, IEEE Trans Syst,
Man, Cybernet, {bf23} (1993), 665-85.
bibitem{1} R. H.
Jones, {it Fitting autoregressions}, J. Amer. Statist. Assoc.,
{bf70(351)} (1975), 590-592.
bibitem{1} M. Khashei, {it
Forecasting the Isfahan Steel Company production price in Tehran
Metals Exchange using Artificial Neural Networks (ANNs)}, Master
of Science Thesis, Isfahan University of Technology, 2005.
bibitem{1} M. Khashei, S. R. Hejazi and M. Bijari, {it A new hybrid
artificial neural networks and fuzzy regression model for time
series forecasting}, Fuzzy Sets and Systems, {bf159} (2008),
bibitem{1} Y. Lin and W. G. Cobourn, {it Fuzzy system
models combined with nonlinear regression for daily ground-level
ozone predictions}, Atmospheric Environment, {bf41} (2007),
bibitem{1} L. Ljung, {it System Identification
Theory for the User, Prentice-Hall}, Englewood Cliffs, NJ, 1987.
bibitem{1} J. T. Luxhoj, J. O. Riis and B. Stensballe, {it A hybrid
econometric-neural network modeling approach for sales
forecasting}, Int. J. Prod. Econ., {bf43} (1996), 175-192.
bibitem{1} S. Makridakis, A. Anderson, R. Carbone, R. Fildes. M.
Hibdon, R. Lewandowski, J. Newton, E. Parzen and R. Winkler, {it The
accuracy of extrapolation (time series) methods: results of a
forecasting competition}, Journal of Forecasting, {bf1} (1982),
bibitem{1} E. Mehdizadeh, S. Sadi-nezhad and R.
Tavakkoli-moghaddam, {it Optimization of fuzzy clustering
criteria by a hybrid pso and fuzzy c-means clustering algorithm},
Iranian Journal of Fuzzy Systems, {bf5} (2008), 1-14
T. Minerva and I. Poli, {it Building ARMA models with genetic
algorithms}, In: Lecture Notes in Computer Science, {bf2037} (2001),
bibitem{1} C. Ong, J. J. Huang and G. H. Tzeng, {it Model
identification of ARIMA family using genetic algorithms}, Appl.
Math. Comput., {bf164(3)} (2005), 885-912.
bibitem{1} P. F. Pai and
C. S. Lin, {it A hybrid ARIMA and support vector machines model in
stock price forecasting}, Omega, {bf33} (2005), 497-505.
bibitem{1} E. Pelikan, C. De Groot and D. Wurtz, {it Power
consumption in West-Bohemia: improved forecasts with decorrelating
connectionist networks}, Neural Network, {bf2} (1992), 701-712.
bibitem{1} M. J. Reid, {it Combining three estimates of gross domestic
product}, Economica, {bf35} (1968), 431-444.
bibitem{1} R.
Shibata, {it Selection of the order of an autoregressive model by
Akaike's information criterion}, Biometrika, {bf AC-63(1)}
(1976), 117-126.
bibitem{1} Z. Tang, C. Almeida and P. A. Fishwick,
{it Time series forecasting using neural networks us,} Box-Jenkins
Methodology Simulation, {bf57(5)} (1991), 303-310.
Z. Tang and P. A. Fishwick, {it Feedforward neural nets as models for
time series forecasting}, ORSA Journal on Computing, {bf5(4)}
(1993), 374-385.
bibitem{1} T. Taskaya and M. C. Casey, {it
A comparative study of autoregressive neural network hybrids},
Neural Networks, {bf18} (2005), 781-789.
bibitem{1} N. Terui and
H. van Dijk, {it Combined forecasts from linear and nonlinear
time series models}, International Journal of Forecasting, {bf18}
(2002), 421-438.
bibitem{1} R. Tsaih, Y. Hsu and C. C. Lai, {it
Forecasting S$&$P 500 stock index futures with a hybrid AI
system}, Decision Support Systems, {bf23} (1998), 161-174.
F. M. Tseng, G. H. Tzeng, H. C. Yu and B. J. C. Yuan, {it Fuzzy ARIMA
model for forecasting the foreign exchange market}, Fuzzy Sets and
Systems, {bf118} (2001), 9-19.
bibitem{1} F. M. Tseng,
H. C. Yu and G. H. Tzeng, {it Combining neural network model with
seasonal time series ARIMA model}, Technological Forecasting $&$
Social Change, {bf69} (2002), 71-87.
bibitem{1} M. V. D. Voort and
M. Dougherty and S. Watson, {it Combining Kohonen maps with ARIMA
time series models to forecast traffic flow}, Transportation
Research Part C: Emerging Technologies, {bf4} (1996), 307-318.
bibitem{1} H. Wold, {it A Study in the analysis of stationary time
series}, Almgrist $&$ Wiksell, Stockholm, 1938.
bibitem{1} H. K. Yu,
{it Weighted fuzzy time-series models for TAIEX forecasting},
Physica A, {bf349} (2004), 609-624.
bibitem{1} L. Yu, S.
Wang and K. K. Lai, {it A novel nonlinear ensemble forecasting model
incorporating GLAR and ANN for foreign exchange rates}, Computers
and Operations Research, {bf32} (2005), 2523-2541.
G. Yule, {it Why do we sometimes get nonsense-correlations
between time series? a study in sampling and the nature of time
series}, J. R. Statist. Soc., {bf89} (1926), 1-64.
G. P. Zhang, {it Time series forecasting using a hybrid ARIMA and
neural network model}, Neurocomputing, {bf50} (2003), 159-175.
bibitem{1} G. Zhang, B. E. Patuwo and M. Y. Hu, {it Forecasting with
artificial neural networks: the state of the art}, International
Journal of Forecasting, {bf14} (1998), 35-62.
bibitem{1} Z. J.
Zhou and C. H. Hu, {it An effective hybrid approach based on grey and
ARMA for forecasting gyro drift, Chaos}, Solitons and Fractals,
{bf35} (2008), 525-529.