Is AR(1)-ARCH(1) covariance stationary?

A_Jo

New member
Joined
Nov 13, 2015
Messages
1
I'm becoming confused by this. Say I have the following model:

\(\displaystyle y_t = c + \phi y_{t-1} + \epsilon_t, \epsilon_{t}|\Omega_{t-1} ~ \tilde{} ~ WN(0,\sigma^2_t),\)
\(\displaystyle \sigma^2_t = \alpha_0 + \alpha_1 + \epsilon^2_{t-1},\)
\(\displaystyle |\phi|<1, \alpha_1<1, \alpha_0 \ge 0, \alpha_1>0.\)

I know that an AR(1) is covariance stationary if \(\displaystyle |\phi|<1\).
I also know that an ARCH(2) is covariance stationary if \(\displaystyle \alpha_0>0, \alpha_1>0\) and \(\displaystyle \alpha_1<1 \).

If those conditions hold does that imply that an AR(1)-ARCH(1) is also covariance stationary?
 
Top