Criterion of Eliminating Single-Parameter Model Uncertainty with Principle of Assuredly Minimal Absolute Lacks and Arithmetic Mean

Автори

There is stated a problem of generation and removal of model uncertainty regarding a single parameter of the being investigated object, which is described with more than one mathematical model. Using the arithmetic mean in removing such single-parameter model uncertainty is compared to the principle of assuredly minimal absolute lacks on the base of the corresponding antagonistic game with symmetric matrix. It has been shown that for the second player optimal strategy, whose support contains equiprobable pure strategies of selecting minimal and maximal values of the being investigated parameter, the corresponding evaluation of the model is not worse than the same evaluation as the arithmetic mean over fixed model values. It is pointed that the nonstrict problem of single-parameter model uncertainty elimination may be solved with the arithmetic mean or principle of assuredly minimal absolute lacks, depending on where minimum of deviate of the being investigated parameter value estimation from its real value is going to be reached. For the strict problem of single-parameter model uncertainty elimination there is suggested to apply all the fixed model values with a probabilistic distribution, being the nearest to the equiprobable distribution within the set of the second player optimal strategies.

Publication year: 
2012
Issue: 
5
УДК: 
519.832.3+519.711.2
С. 75—80. Бібліогр.: 13 назв.
References: 

1. Трухаев Р.И. Модели принятия решений в условиях неопределенности. — М.: Наука, 1981. — 258 с.
2. Черноруцкий И.Г. Методы принятия решений. — СПб.: БХВ-Петербург, 2005. — 416 с.
3. Воробьев Н.Н. Основы теории игр. Бескоалиционные игры. — М.: Наука, Глав. ред. физ.-мат. лит-ры, 1984. — 496 с.
4. Воробьев Н.Н. Теория игр для экономистов-кибернетиков. — М.: Наука, Глав. ред. физ.-мат. лит-ры, 1985. — 272 с.
5. S. de Wit and G. Augenbroe, “Analysis of uncertainty in building design evaluations and its implications”, Energy and Buildings, vol. 34, is. 9, pp. 951—958, 2002.
6. C.J. Hopfe and J.L.M. Hensen, “Uncertainty analysis in building performance simulation for design support”, Ibid, vol. 43, is. 10, pp. 2798—2805, 2011.
7. A. Smith et al., “Analysis of a combined cooling, heating, and power system model under different operating strategies with input and model data uncertainty”, Ibid, vol. 42, is. 11, pp. 2231—2240, 2010.
8. S. Andersson et al., “A random wear model for the interaction between a rough and a smooth surface”, Wear, vol. 264, is. 9-10, pp. 763—769, 2008.
9. T. Nilsen and T. Aven, “Models and model uncertainty in the context of risk analysis”, Reliability Engineering & System Safety, vol. 79, is. 3, pp. 309—317, 2003.
10. I. Park et al., “A Bayesian approach for quantification of model uncertainty”, Ibid, vol. 95, is. 7, pp. 777—785, 2010.
11. J. Jacques et al., “Sensitivity analysis in presence of model uncertainty and correlated inputs”, Ibid, vol. 91, is. 10, 11, pp. 1126—1134, 2006.
12. Романюк В.В. Метод реалізації оптимальних змішаних стратегій у матричній грі з порожньою множиною сідлових точок у чистих стратегіях з відомою кількістю партій гри // Наукові вісті НТУУ “КПІ”. — 2009. — № 2. — С. 45—52.
13. Романюк В.В. Обчислювальний метод реалізації оптимальної змішаної стратегії у матричних іграх // XV Int. Conf. “Problems of Decision Making Under Uncertainties (PDMU-2010)”, May 17—21, 2010, Lviv: abstracts. — Lviv, 2010. — P. 142—144.

AttachmentSize
2012-5-17.pdf211.06 KB

Тематичні розділи журналу

,