in Forecasting: Principles and Practice there's a warning that

inverse characteristic roots close to the unit circle may be numerically unstable, and the corresponding model will not be good for forecasting

Is this the only way to discover this instability? or are there other telltale signs such as the s.e. of the ar/ma terms being large, or the confint being large or including zero? or is there no relation?


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.