To derive the log likelihood function of an MA(1) we condition on \(\displaystyle \epsilon_0 \).
But when deriving the log likelihood function function for an AR(1) we don't.
Why not? Is it just because \(\displaystyle \epsilon_0 \) isn't found in the AR(1) equation?
But when deriving the log likelihood function function for an AR(1) we don't.
Why not? Is it just because \(\displaystyle \epsilon_0 \) isn't found in the AR(1) equation?