In this answer of mine there is a disagreement with another user over the validity of my answer.

In response to the user's comment, I edited my answer to provide an explanation of why they were wrong along with an empirical example to demonstrate my point.

The point in discussion is whether a Seasonal ARIMA model can somehow go out of phase over time or not (by definition it can't since the seasonality is built into the model - but the other user doesn't seem to understand that). The user also seems to be making the outlandish suggestion that forecasting a monthly time series 17 years into the future doesn't constitute a long forecast horizon.

The user kept coming back to the point and arguing for his point of view. I've given up trying to debate the issue with him.

But now he's taken it a step further and edited my answer, adding a disclaimer that it is disputed.

I don't want to get into an edit/counter edit battle with this guy (especially since he has a much higher reputation score than me). At the same time I'm certain about my assertion with regards to Seasonal ARIMA models and don't want other users to be misled by the edits and comments.

What is the policy on this? How should I proceed?


Substantive content should not be changed in someone else's answer. Nor should editorial content be added. People who disagree should leave comments, downvote, and/or post their own answer. In their own answer, they can call out the other answer and explain what they disagree with and why.

At all times all parties should abide by SE's be nice policy. If you believe that is violated, or that someone is editing inappropriately, flag it for moderator attention. Don't get involved in a flame war.

No. Reasons for editing are described in here:

When should I edit posts?

Any time you feel you can make the post better, and are inclined to do so. Editing is encouraged!

Some common reasons to edit are:

  • to fix grammatical or spelling mistakes
  • to clarify the meaning of a post without changing it
  • to correct minor mistakes or add addendums / updates as the post ages
  • to add related resources or hyperlinks

Tiny, trivial edits are discouraged - try to make the post significantly better when you edit, correcting all problems that you observe.

It's neither full, nor definite explanation, but it describes the general idea about editing: it is about minor improvements in form, not the content, or meaning of the question or answer that is edited. The description mentions correcting "mistakes", but in such cases it is usually better just to comment pointing the mistake and suggesting edit.

On another hand, there are cases where you could probably edit the question to correct the mistake (e.g. correct a minor arithmetic error in the calculation, if you're 100% sure and carefully checked your own calculation!), but in such cases you should always additionally comment the answer and describe what and why did you edit.

The only case where we are allowed to edit the content is community Wiki threads, where anyone can make edits in answers, since they are maintained by the community, not the particular users:

Community Wiki posts work by partly transferring ownership of the post from the original author to the community. They make the post easier to edit and maintain by a wider group of users, but they do not contribute to any user's reputation.

If you disagree with the answer, you can always downvote and/or comment it, it it is inappropriate for the site ($\ne$ incorrect), just flag it for moderator attention.

FWIW, I deliberately have not read the thread in question and therefore have no views here on who is more nearly right, and so forth.

I think this is more general inter-personal etiquette than site policy.

It's OK in my view in most cases to correct small presentation errors in anyone's posts. That is especially easy if you feel you know the other party well in CV terms and there's good mutual regard. (If you're in doubt about that, it doesn't apply.)

But I would not go there if I was strongly disagreeing with someone technically and/or there was also a history of the person taking criticism badly or otherwise behaving poorly (e.g. a known record of suspensions, angry or bitter exchanges, downvoting etc.).

Any reputation-based right to make edits doesn't over-rule a need to maintain courtesy and sensitivity. Even minor corrections can be misread as pointed signals (and you can't even spell, punctuate, choose notation carefully, understand basic probability, ...).

The way to disagree, as others are pointing out, is through comments and your own answer (which can cross-refer to other answers you disagree with).

Meta-meta: I edited your question for a minor typo. That was nothing to do with your question or argument.

Your main point is well answered already. I will just address what you could do about the technical argument leading to disagreement: Can a seasonal ARIMA model mimic a periodic model? Can you please add that as a question (with links back to your answer leading to this disagreement), so it can be addressed and discussed properly?

  • 5
    (+1) This is a good way to leverage the mechanisms of CV to start productive discussions. – Sycorax Aug 3 at 18:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .