Edge Rewrite
// HTMLRewriter · presentation

This page was redesigned at the edge.

Cloudflare fetched the original article and streamed it through HTMLRewriter to apply an entirely new visual system without rebuilding the source page.

// request.cf · coarse context

A page that knows where it met you.

Only coarse request metadata is shown. This demo does not display or persist visitor IP addresses.

Country
US
Cloudflare location
CMH
Connection
HTTP/2
Language
Not provided

Ray ID: a2819bfec9e297b7

Jump to content

Talk:Commodity channel index

Page contents not supported in other languages.
Add topic
From Wikipedia, the free encyclopedia
Latest comment: 16 years ago by Slawkenbergius in topic Interpretation

Normal deviation

[edit]

The term "normal deviation" is non-standard in statistics. If it should be "standard deviation," make it so.74.70.32.60 18:00, 22 January 2007 (UTC)Reply

It's not a starndard deviation.

[edit]

No, it doesn't use a standard deviation-> sqrt(average(error^2))

It uses this-> sqrt(average(abs(error)))

maybe we should just change the symbol used there.

Sukisuki 15:18, 14 November 2007 (UTC)Reply

It uses the mean absolute deviation, by design. —Preceding unsigned comment added by 98.183.211.225 (talk) 18:31, 30 June 2009 (UTC)Reply

Presentation

[edit]

I've always like the aesthetics of the brolinger band, that it's drawn right on top of the chart itself... to me at least, CCI would be much more meaningful if people did that. if you dissect it a little it's easy you see that it's the same idea as a brolinger band, but with three changes: using that 'typical price' instead of period averages; using an average of the absolute value of first power of the error instead of the second to calculate the 'deviation'; always using 1.5 'deviations'. Sukisuki 15:28, 14 November 2007 (UTC)Reply

Interpretation

[edit]

There are interpretations in the Calculation section which seem to differ with the interpretations in the Interpretation Section.

--Slawkenbergius (talk) 17:12, 27 January 2010 (UTC)Reply