The Wold representation says that any stationary process is equivalent to a MA($\infty$) process which, in turn, is the limit of a MA(q) process. Does that mean that anything that can be proved for an MA(q) immediately generalizes to any stationary process? I suspect this is not the case. but can't really come up with a counter-example.

  • 2
    $\begingroup$ For the process to be stationary there are conditions on the moving average parameters. $\endgroup$ – Michael R. Chernick Feb 6 '17 at 22:26

Proposition. Let $y_t$ be a series from $MA(q)$. Then for any $q$, there is $s$ such that $Cov(y_t,y_{t-s})=0$.

Obviously, the above proposition is true. But there is no such $s$ for $AR(1)$ process with non-zero AR coefficient.

| cite | improve this answer | |
  • $\begingroup$ Nice example, but you did not make any conclusion. Since it was not self-study, it makes sense to spell it out fully. $\endgroup$ – Richard Hardy Feb 7 '17 at 6:36
  • $\begingroup$ What conclusion? I provided an counter-example to the OP's question, which means that his hypothesis is false. $\endgroup$ – Julius Feb 7 '17 at 6:46
  • $\begingroup$ That's what I meant. This sentence was missing :) $\endgroup$ – Richard Hardy Feb 7 '17 at 7:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.