We have seen what the m-step forecast for an AR(2) process is. We will now look at the error of this estimate. In all cases, assume the available path length n is sufficiently long (i.e. greater than the order of the AR process).

a. What is the variance of the 1-step forecast for AR(2)?

b. What is the variance of the 2-step forecast for AR(2)?

c. What is the variance of the 3-step forecast for AR(2)?

(Bonus 5pts) What is the variance of the 2-step forecast for AR(m), m > 3?

We have seen what the m-step forecast for an AR(2) process is. We will now look at the error of this estimate. In all cases, assume the available path length n is sufficiently long (i.e. greater than the order of the AR process). 2 a. What is the variance of the 1-step forecast for AR(2)? b. What is the variance of the 2-step forecast for AR(2)? c. What is the variance of the 3-step forecast for AR(2)? 5pts) What is the variance of the 2-step forecast for AR(m), m > 3? We have seen what the m-step forecast for an AR(2) process is. We will now look at the error of this estimate. In all cases, assume the available path length n is sufficiently long (i.e. greater than the order of the AR process). 2 a. What is the variance of the 1-step forecast for AR(2)? b. What is the variance of the 2-step forecast for AR(2)? c. What is the variance of the 3-step forecast for AR(2)? 5pts) What is the variance of the 2-step forecast for AR(m), m > 3