And for example, let's take SMA(2) model

in this table does there exist white noise ? Which value I observe to decide the existance of white noise? Please explain it. Thank you

enter image description here

  • $\begingroup$ Hint: white noise is about absence - rather than presence - of regularities. Also, your estimated MA process is non-invertible and thus would produce explosive behaviour into the future; does that make sense for the data you are studying? Do you expect your left-hand-side variable to "explode" in the future? $\endgroup$ – Richard Hardy Mar 2 '15 at 19:23
  • $\begingroup$ okay thank you. what i understand is that the p-value of MA(8) 0.3065 > 0.05 ($\alpha = 0.05$) thus there exists white noise.it's not suitable model. @RichardHardy Right? $\endgroup$ – user315 Mar 2 '15 at 19:44
  • $\begingroup$ White noise has very little to do with one of the coefficients in the model being not statistically significant... You will not learn much by making wild guesses. Reading a textbook could be wiser. $\endgroup$ – Richard Hardy Mar 2 '15 at 19:54
  • $\begingroup$ well, can you suggest a best book to learn them ? @RichardHardy $\endgroup$ – user315 Mar 2 '15 at 19:55
  • 1
    $\begingroup$ These are not necessarily the best ones. As I said, I have forgotten what I used to like. There are so many introductory econometrics textbooks, so if you do not like Wooldridge or Zivot & Wang, you will find many more to choose from, just look around. $\endgroup$ – Richard Hardy Mar 2 '15 at 20:12

Your Answer

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

Browse other questions tagged or ask your own question.