Title :
Modeling of the Uncertainties at Conditional Heteroskedastic (ARCH)
Author :
Shamsha, Boris ; Khalina, Iuliia ; Shatovskaya, Tatyana
Author_Institution :
Kharkiv Nat. Univ. of Radio Electron., Kharkiv
Abstract :
The important rule of modern economic design consists in the comprehensive checking of the estimated model for the article of violation of one or another suppositions. In work some of methods of diagnostics of such violations in the context of regressive analysis are transferred and it is told about that of them, which was utilized in the applied part of work at the design of inflation: criterion of autocorrelation of tailings. Criterion of functional form (RESET), criterion of geteroskedastic, criterion of normality. The index of influentialness (DFFITS) is considered also and to, what modification it follows to subject the coefficient of determinatsii in models with determined by trends. Inconstancy of mechanisms, insufficiency of set of data, false regressions can create serious problems for an economic design. Although fully will manage it is impossible with problems, but it is important to realize that problems exist and carefully behaves to the results got at a design.
Keywords :
autoregressive processes; inflation (monetary); regression analysis; auto regressive conditionally heteroskedastic; economic design; financial market; inflation design; regressive analysis; Autocorrelation; Consumer electronics; Decision making; Fractals; Globalization; Information processing; Information technology; Instruments; Reactive power; Uncertainty; Value-at-Risk; fractal; geteroskedastic; market risk; volatility;
Conference_Titel :
CAD Systems in Microelectronics, 2007. CADSM '07. 9th International Conference - The Experience of Designing and Applications of
Conference_Location :
Lviv-Polyana
Print_ISBN :
966-533-587-0
DOI :
10.1109/CADSM.2007.4297585