AR(1)-GARCH(1,1)

[scottyby23]

Hey everyone, new to this forum today! Need some help with my GARCH model.

I have been told to use an AR(1) specification in my mean equation and to estimate a GARCH model. I have the following output. I do however notice that in my mean equation, both coefficients are insignificant. What are the implications of this result? I am quite stumped about what this exactly means, given that the coefficients in the variance equation are significant.

Thanks for any help!
Bob


Dependent Variable: X
Method: ML - ARCH (Marquardt) - Normal distribution
Sample (adjusted): 1/11/1988 5/20/2011
Included observations: 6095 after adjustments
Convergence achieved after 23 iterations
Presample variance: backcast (parameter = 0.7)
GARCH = C(3) + C(4)*RESID(-1)^2 + C(5)*GARCH(-1)

Variable Coefficient Std. Error z-Statistic Prob.

C 0.000253 0.000231 1.096162 0.2730
X(-1) -0.018717 0.012800 -1.462288 0.1437

Variance Equation

C 4.35E-06 4.59E-07 9.495379 0.0000
RESID(-1)^2 0.062211 0.002846 21.85986 0.0000
GARCH(-1) 0.933636 0.002301 405.8217 0.0000


R-squared 0.001479 Mean dependent var 0.000287
Adjusted R-squared 0.001315 S.D. dependent var 0.026018
S.E. of regression 0.026001 Akaike info criterion -4.776462
Sum squared resid 4.119131 Schwarz criterion -4.770954
Log likelihood 14561.27 Hannan-Quinn criter. -4.774551
Durbin-Watson stat 2.061818