Talk:Moving average/Archives/2017

Comments
The introduction to the article seems to assume that the reader already knows the material that the article imparts. The body of the article is comprehensible, but not easily. 64.26.98.90 20:41, 8 November 2007 (UTC)

Perhaps this is technical but it is about the only place I was able to find enough detail to actually code and verify my EMA. I appreciate the author(s) very much! Well done! Brittlowry (talk) 22:52, 1 January 2008 (UTC)

The second paragraph uses the word 'nuparticular'. A quick google search for "nuparticular meaning" provides this wiki page as the first result. Perhaps another word should be chosen? — Preceding unsigned comment added by 108.15.31.21 (talk) 20:02, 13 March 2017 (UTC)

What is meant by, "If the data used are not centered around the mean, a simple moving average lags behind the latest datum point by half the sample width." This could use clarification and some explanation as to why. — Preceding unsigned comment added by 108.15.31.21 (talk) 21:15, 13 March 2017 (UTC)

Article rollback
I reverted the article to the version before User:Bert Niehaus began a long series of (good faith) edits. It wound up undoing a few edits interspersed among those, but it looked like those were all just related to fixing introduced mistakes, so hopefully that's not a problem.

Some of these edits may have been helpful, but there were simply too many go through each one individually. Specifically, there was an awful lot of original research involved. There were also a lot of English mistakes (granted, it's still better than I could have done with another language), and it would be helpful to have a fluent speaker look over the changes to make.

Unfortunately, a lot of the math was questionable too, such as the attempt to generalize to arbitrary groups -- the exposition just didn't make any sense.

If someone (original editor included) wishes to reintroduce any of these changes, it might be a good idea to discuss here as well. --Deacon Vorbis (talk) 13:58, 30 April 2017 (UTC)