I have found an article about stationarity: Variance of a stationary AR(2) model
and I have also estimated a model:
And I am not sure if I have understood it, is it truly stationary? And if so, could you show me how to derive it analytically?
Asked
Active
Viewed 81 times
1
-
1Your estimates assume a stationary model. What exactly do you wish to "derive analytically"?? – whuber Jun 29 '22 at 19:23
-
@whuber What I mean is, I have heard that the AR coefficients cannot be equal nor bigger than 1, and as far as I know there are some exceptions from it. The problem is, I'm not sure how to prove it while having an actual example. – Fatafim Jun 29 '22 at 20:41
-
1Check out relevant posts about "unit roots" of AR processes. For a fuller appreciation, one subtlety is treated at https://stats.stackexchange.com/questions/406204. – whuber Jun 29 '22 at 20:52
1 Answers
1
You can write the process in companion form (as a VAR(1)): $$\left[ \begin{array}{c} x_{n} \\ x_{n-1}% \end{array}% \right] =\left[ \begin{array}{c} 2.84 \\ 0% \end{array}% \right] +\left[ \begin{array}{cc} 1.36 & -0.75 \\ 1 & 0% \end{array}% \right] \left[ \begin{array}{c} x_{n-1} \\ x_{n-2}% \end{array}% \right] +\left[ \begin{array}{c} \varepsilon _{n} \\ 0% \end{array}% \right] $$ The eigenvalues of the autoregression matrix are within the unit circle. Therefore, the process is stationary.
user4422
- 1,178
-
What is the autoregression matrix? The one with elements 1.36, -0.75, 1 and 0? – Richard Hardy Jun 30 '22 at 08:51