On the parameters identification in models of assets dynamics
Abstract
The problem of identification of parameters of mathematical models of dynamic processes is considered that can be described by ordinary differential equations and systems of such equations. Using an example of mathematical models of dynamic formation of the market value per share and stock portfolio, algorithms were developed for constructing the optimal values of parameters of such models. The algorithms for parametric identification and optimization are based on iterative procedures that allow, at each step, to obtain "the best" values of model’s parameters based on selected quality criteria. Guaranteed parameters’ estimations are built in the class of ellipsoidal sets that allows to obtain the guaranteed financial performance of investment operations using the mathematical problems of the financial analysis as an example.References
Garashchenko F.G., Kulyan V.R., Rutitskaya V.V. Modelirovaniye i analiz dinamiki investitsiy // Problemy upravleniya i informatiki. – 2001. – № 6. – S.109–119.
Garashchenko F.H., Kulyan V.R., Rutyts'ka V.V. Zastosuvannya metodiv praktychnoyi stiykosti dlya rozv"yazuvannya zadach investytsiynoho menedzhmentu // Visnyk Kyyivs'koho universytetu. Seriya: fizyko-matematychni nauky. – 2005. – # 3. – S. 232–239.
Garashchenko F., Kulian V., Rutitskaya V. Modelling and Analysis of Investment Trends // Journal of Automation and Information. – 2011. – 43, issue 12. – P. 8–58.