Decision support system for forecasting financial processes on the basis of system analysis principles
Keywords:financial processes, uncertainty, system analysis, adaptation, forecasting
AbstractA concept is proposed for solving the problem of adaptive forecasting that is based on the system analysis methodology and combined use of preliminary data processing techniques, mathematical and statistical modeling, forecasting and optimal state estimation of the processes under study. The cyclical adaptation of a structure and model parameters on the basis of a set of statistical characteristics of a process under study provides a possibility for reaching high quality estimates of forecasts under condition that data is informative. To identify and take into consideration possible stochastic, structural and parametric uncertainties it is proposed to use optimal and digital filtering and data mining methods such as Bayesian networks, adaptive BN, particle filter and other instruments. Possible parametric uncertainties are minimized with application of several alternative parameter estimation techniques such as LS, RLS, ML and Markov chains Monte Carlo sampling. The conducted study suggests that the proposed methodology can be applied to the analysis of a wide class of real life processes including nonlinear nonstationary processes in finances, economy, ecology and demography.
Holsapple C.W. Decision support systems / C.W. Holsapple, A.B. Winston. — Saint Paul (USA): West Publishing Company, 1996. — 850 p.
Turban E. Decision support systems / E. Turban, J.E. Aronson. — New Jersey: Prentice Hall, 2001. — 865 p.
Lukashin Ju.P. Adaptivnye metody kratkosrochnogo prognozirovanija / Ju.P. Lukashin. — M.: Finansy i statistika, 2003. — 414 s.
Bidjuk P.I. Analiz chasovykh rjadiv / P.I. Bidjuk, V.D. Romanenko, O.L. Tymoshchuk. — K.: NTUU "KPI", 2013. — 600 s.
Press S.J. Subjective and objective Bayesian statistics / S.J. Press. — Hoboken (New Jersey): John Wiley & Sons, Inc., 2013. — 560 p.
Rossi P.E. Bayesian statistics and marketing / P.E. Rossi, G.M. Allenby, R. McCulloch. — New Jersey: John Wiley & Sons, Ltd, 2005. — 348 p.
Diebold F.X. Forecasting / F.X. Diebold. — Pennsylvania: University of Pennsylvania, 2018. — 800 p.
Zgurovskij M.Z. Analiticheskie metody kalmanovskoj fil'tratsii / M.Z. Zgurovskij, V.N. Podladchikov. — K.: Nauk. dumka, 1995. — 285 s.
Ng B.M. Adaptive dynamic Bayesian networks / B.M. Ng // Joint Statistical Meetings, Salt Lake City (USA), July 29 – August 2, 2007. — P. 1–7.
Zgurovsky M.Z. Method of constructing Bayesian networks based on scoring functions / M.Z. Zgurovsky, P.I. Bidyuk, O.M. Terentyev // Cybernetics and System Analysis. — 2008. — Vol. 44, N 2. — P. 219–224.
http://www.mataf.net/en/tools/home / Database.
Nong Y. The Handbook of Data Mining / Y. Nong. — New Jersey: Arizona State University Publishers, 2003. — 1201 p.
Altman E.I. Application of Classification Techniques in Business, Banking and Finance / E.I. Altman, R.B. Avery, R.A. Eisenbeis, J. Sinkey. — Greenwich: JAI Press, 1981. — 418 p.
Hosmer D.W. Applied Logistic Regression / D.W. Hosmer, S. Lemeshow. — New York: John Wiley & Sons, Inc., 2000. — 380 p.
Cowell R.G. Probabilistic networks and expert systems / R.G. Cowell, A.P. Dawid, S.L. Lauritzen, D.J. Spiegelhalter. — New York: Springer, 1999. — 323 p.