Advances and Applications in Statistics
Volume 7, Issue 3, Pages 403 - 416
(December 2007)
|
|
MIXED GARCH-JUMP MODELS WITH GENERALIZED ERROR DISTRIBUTION FOR ASSETS RETURNS
Yu-Jan Shen (Taiwan, R. O. C.), Kuan-Fu Shen (Taiwan, R. O. C.) and Ming-Chih Lee (Taiwan, R. O. C.)
|
Abstract: In most financial asset returns, the most commonly used parametric specification for the return distribution is the standard normal distribution and discrete jumps in returns for observed data. In this paper, we use the daily return from the DJIA to show that the assumption of generalized error distribution (GED) which takes the normal distribution as a special case. Furthermore, we are the first one to propose the concept of GARJI-GED model which extends the GARJI models in Chan and Maheu [11]. We show that the GARJI-GED model specification provides a significant statistical improvement over the assumption of normal return distribution model. That is, the GED return distribution reaches the assumption of financial asset returns distribution in GARJI model specification. |
Keywords and phrases: DJIA, GARJI, generalized error distribution (GED), jump intensity, normal distribution. |
|
Number of Downloads: 351 | Number of Views: 1312 |
|