تقدير معلمات أنموذج Garch لتوزيع القاطع الزائدي مع التطبيق

Abstract EN

The study of AutoRegressive Conditional Heteroscedasticity modelsof the important studies in time series, especially financial ones, becausemost of the financial time series have high Volatility that is, theconditional variance in them is not fixed and depends on the past, thatmost of the previous studies rely on some distributions such as normaldistribution GED distribution and t distribution.

In this paper, a new distribution of the GARCH model, which ishyperbolic- secant distribution, is presented in theory and practice.

Asample representing the opening price of the Brent oil was found and thedata test showed that it follows the hyperbolic secant distribution.

Aftertesting it was found to be a problem for ARCH.

After estimating it wasfound that the best model to represent this data according to the best testcriteria is AIC, SIC, H-Q, GARCH (1, 1).

Key Words: ARCH, Brent Oil Prices, Hyperbolic Secant Distribution, Model selection