Możliwości i problemy wykorzystania sieci neuronowych w prognozowaniu procesów gospodarczych

• Autorzy: Ryszard Tadeusiewicz

Warianty tytułu

Potentials and Problems Underlying Application of Neuron Networks for Prognosticating Economic Trends

• Języki publikacji: PL

Abstrakty

W pracy omówiono zagadnienie nowych możliwości, jakie stwarza technika sieci neuronowych w zadaniach związanych z prognozowaniem procesów gospodarczych. Zasygnalizowano w niej jedynie fakt, że z wielu badań i eksperymentów wynika bezsporna przydatność sieci neuronowych jako narzędzi służących prognozowaniu ekonomicznemu - np. kursów walut, ceny akcji czy wiarygodności kredytobiorców. Ze względu na ogromną liczbę publikacji, jakie ukazały się na ten temat w ostatnich latach, w pracy zebrano i omówiono w skrócie jedynie najważniejsze wyniki. Praca nie może więc być traktowana jako wyczerpujące kompendium obejmujące wszystkie aspekty wymienionego w tytule zagadnienia, należy jednak mieć nadzieję, że wskazując na istniejące możliwości i odsyłając do konkretnych danych zawartych w bogatej literaturze przedmiotu - praca spełni swoje zadanie, inspirując polskich badaczy do podjęcia samodzielnych prac w zakresie zagadnień prognozowania i wykorzystaniem sieci neuronowych, a praktyków (zwłaszcza organizatorów życia gospodarczego) zachęci do sięgania po nowe, ale w wielu przypadkach wysoce użyteczne narzędzie, jakim jest technika sieci neuronowych. (abstrakt oryginalny)

EN

The paper takes up the issue of new potentials generated by neuron networks in the accomplishment of tasks connected with the prognostication of economic trends. Because of the obvious limitations of size the paper only outlines the fact that based on the numerous research and experiments conducted so far it becomes unambiguously clear that neuron networks come as a useful tool of economic projections involving currency rates, stock prices or customer credit worthiness. Because of the huge number of publications which have recently become available on the market the author has collected and dealt with only the most significant of those and even there has done so briefly. The paper can by no means be treated as a comprehensive manual embracing all aspects of the issue named in the title, but it should only to be hoped by pointing out potentials and referring to specific data which are to be found in the ample literature dealing with the subject that it will meet its objective by inspiring Polish researchers and scientists to do their independent research in the field of neuron network based research on the one hand and will encourage practitioners especially those in charge of the economy to reach for new tools, and in many cases very useful ones, on the other, such as the technology of neuron networks. (original abstract)

Słowa kluczowe

PL

Prognozy gospodarcze; Sieci neuronowe

EN

Economic forecast; Neural networks

• Czasopismo: Zeszyty Naukowe / Akademia Ekonomiczna w Krakowie
• Rocznik: 1997
• Numer: nr 493
• Strony: 5--18

Twórcy

autor

Ryszard Tadeusiewicz

Bibliografia

• Back A.D., Ah Chung Tsoi [1992], An Adaptive Lattice Architecture for Dynamic Multilayer Perceptions, Neural Computation, Nov., Vol. 4, lss. 6.
• Benaim ML, Tomasini L. [1992], Approximating Functions and Predicting Time Series with Multi-sigmoidal Basis Functions, Artificial Neural Networks 2, Proceedings of the 1992 International Conference (ICANN-92), Elsevier, Amsterdam Netherlands, Vol. 1.
• Chang-Ching Lin, Hsu-Pin Wang [1993], Classification of Autoregressive Spectral Estimated Signal Patterns Using an Adaptive Resonance Theory Neural Network, Computers in Industry, Aug., Vol. 22, iss. 2.
• Davenport M.R., Day S.P. [1992], Chaotic Signal Emulation Using a Recurrent Time Delay Neural Network, Neural Networks for Signal Processing SI, Proceedings of the IEEE-SP Workshop, IEEE, New York.
• DeMers D. [1992], Dimensionality Reduction for Nonlinear Time Series, Proceedings of the SPIE - The International Society for Optical Engineering, Vol. 1766.
• Developing Tools and Methods for Applications Incorporating Neitro Fuzzy and Chaos Technology [1993], R. Katayama, Y. Kajitani, K. Kuwata, Y. Nishida, Computers & Industrial Engineering, Oct., Vol. 24, lss. 4.
• Hentschel H.G.E., Jiang Z. [1993a], Learning to Control Dynamical Behaviour, Physica D, Aug., Vol. 67, iss. 1-3.
• Hentschel H.G.E., Jiang Z. [1993b], Prediction Using Unsupervised Learning, Physica D, Aug., Vol. 67, Iss. 1-3.
• Hsu W.,Tenorio M.F. [1993], Plastic Network for Predicting the Mackey-Glass Time Series [w:] Third Workshop on Neural Networks, Soc. Comput. Simulation, San Diego CA USA.
• Ikonomopoulos A., Tsoukalas L.H., Uhrig R.E. [1993], Integration of Neural Networks with Fuzzy Reasoning for Measuring Operational Parameters in a Nuclear Reactor, Nuclear Technology, Oct., Vol. 104, Iss. 1.
• Intelligent Stock Trading System with Price Trend Prediction and Reversal Recognition Using Dual-module Neural Networks [1993], J. Gia-Shuh, L. Feipei, J. Bor-Wei, P. Tai-Ming, Ch. Li-Hua, Applied Intelligence: The International Journal of Artificial Intelligence Neural Networks and Complex Problem-Solving Technologies, Sept., Vol. 3, Iss. 3.
• Kondo T. [1993], Short-term Prediction of Air Pollution Concentration by a Neural Network, Transactions of the Society of Instrument and Control Engineers, June, Vol. 29, Iss. 6.
• Kuo J.-M., Principle J.C., Vries de B. [1992], Prediction of Chaotic Time Series Using Recurrent Neural Networks, Neural Networks for Signal Processing II, Proceedings of the IEEE-SP Workshop, IEEE, New York.
• Leung H., Haykin S. [1993], Rational Function Neural Network, Neural Computation, Nov., Vol. 5, Iss. 6.
• Littmann E., Rittcr ll. [1992], Cascade LLM Networks, Artificial Neural Networks 2, Proceedings of the 1992 International Conference (ICANN-92), Vol. 1, Elsevier, Amsterdam.
• Lohrbach T., Schumann M. [1992], Artificial Neural Networks and ARIMA-Models within the Field of Stock Market Prediction - a Comparison, Adaptive Intelligent Systems, Proceedings of the BANKA1 Workshop, Elsevier, Amsterdam.
• Martinez T.M., Berkovich S.G., Schulten K.J. [1993], "Neural-gas" Network for Vector Quantization and Its Application to Time-series Prediction, IEEE Transactions on Neural Networks, July, Vol. 4, Iss. 4.
• Next Day Peak Load Forecasting Using a Multilayer Neural Network with an Additional Learning [1993], Y. Morioka, K. Sakurai, A. Yokoyamu, Y. Sekine, w: Y. Tumurj H. Suzuki, H. Mori (eds), Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems, New York.
• Pethel S.D., Bowden C.M. [1992], Neural Network Applications to Nonlinear Time Series Analysis, Proceedings of the 1992 International Conference on Industrial Electronics Control Instrumentation and Automation, Power Electronics and Motion Control, Vol. 2, IEEE, New York.
• Raich A., Wu X., Cinar A. [1993], Approximate Dynamic Models for Chemical Processes: A Comparative Study of Neural Networks and. Nonlinear Time Series Modeling Techniques Dynamics and Control of Chemical Reactors Distillation Columns and Batch Processes Selected Papers from the 3rd IFAC Symposium, Balchen J.G. Pergamon Press, Oxford.
• Rao S.S., Sethuraman S., Ramamurti V. [1992], A Recurrent Neural Network for Nonlinear Time Series Prediction - a Comparative Study, Neural Networks for Signal Processing n Proceedings of the IEEE-SP Workshop, IEEE, New York.
• Schmidhuber J., Prelinger D. [1993], Discovering Predictable Classifications, Neural Computation, July, Vol. 5, Iss. 4.
• Taylor J.G. [1992], Temporal Sequence Storage, Artificial Neural Networks 2. Proceedings of the 1992 International Conference ICANN-92, vol. 2, Amsterdam.
• Times Series and Neural Networks: A Statistical Method for Weight Elimination [1993], M. Cottrell, B. Girard, Y. Girard, M. Mangeas, w: European Symposium on Artificial Neural Networks ESANN, Brussels.
• Toda N., Murai N., Usui S. [1992], A Measure of Nonlinearity in Time Series Using Neural. Network Prediction Model, Artificial Neural Networks 2, Proceedings of the 1992 International Conference ICANN-92, Vol. 2, Amsterdam.
• Wang Y.J., Kubik K.K. [1992], A Neural Network Approach to Anti-SA modelling for GPS Users, ISSPA 92, w: Proceedings of Third International Symposium on Signal Processing and its Applications, Vol. 2, D.A. Gray, IREE, Edgecliff NSW, Australia.
• Wulff N.H., Hertz J.A. [1992], Prediction with Recurrent Networks, Neural Networks for Signal Processing II, Proceedings of the IEEE-SP Workshop, IEEE, New York.