Talk:Extrapolation

From Wikipedia, the free encyclopedia

extarpolation

Is that "to remove a tarp from"? scot 20:21, 4 Apr 2005 (UTC)

Contents 1 Recent additions issues 2 COMPUTERISED FRENCH CURVES 3 Extrapoltion a form of interpolation 4 Contrast Extrapolation with Interpolation 5 Removing statistics content

[edit] Recent additions issues

To: 205.162.232.230

Thank you for your additions to extrapolation. I would like to mention however, that you chose to talk about a very narrow type of extapolation, that being in the complex plane. And that was not made clear. I think your contributions need some more editing. Would you mind making yourself an account, so that we can keep in touch with you, and do more work on this? Thanks. Oleg Alexandrov 20:56, 4 Apr 2005 (UTC)

I prodded him into adding this entry, so I'll pass the request on. scot 21:00, 4 Apr 2005 (UTC)

Prodded person: I'll think about it.

Prodded person: AFAICT, all of my edits to the Wikipedia are of the form, "write something half-formed that irritates everyone else to fill in the rest of the article." This sort of speaks against having a real account.

Sure, he turns me into a Wikiholic (he requested "original Wikipedia entries" for his birthday last year, which led me to create internal, external, and terminal ballstics) and he doesn't even have an account... scot 21:05, 4 Apr 2005 (UTC)

I found the above quite amusing. However, we cannot accept half-done stuff on Wikipedia. I might consider reverting the new contributions if they are not in good shape, as we'd better have a stub which might become a good article in due time than having a contribution hard to understand and hard to modify by others. Oleg Alexandrov 21:08, 4 Apr 2005 (UTC)

[edit] COMPUTERISED FRENCH CURVES

This method is described in detail in www.AIDSCJDUK.info It does not appear to be in any mathematical treatise. However, it has been giving accurate pojections of AIDS/HIV data in the UK since 1987. It suits any distribution with a tendency to be exponential biut with limiting or accelerating factors. Edwardhfd@aol.com

(I moved the above here from the main article. StuRat 22:00, 7 October 2005 (UTC))

[edit] Extrapoltion a form of interpolation

Does anyone care to justify this statement? It has been in the article since it was created (anonymously).

It seems to me to be nonsense to describe extrapolation as a form of interpolation. Certainly the two are generalisations of something wider, but I would not describe either as a form of the other. I will write a more concise description and replace the current one when I get a round tuit if no one objects. 81.111.34.148 14:23, 3 November 2005 (UTC)

I agree with you there. Also, please sign up so we can get to know you better. This also serves another purpose, as contributions from anonymous I/Ps are frequently vandalism, so we must spend more time checking them. StuRat 14:32, 3 November 2005 (UTC)

Sorry, session timed out I think. I am User:TomViza. TomViza 15:49, 3 November 2005 (UTC)

The connection between extrapolation and interpolation is now mentioned in the section on linear extrapolation. --Berland 16:02, 24 June 2007 (UTC)

[edit] Contrast Extrapolation with Interpolation

we can contrast the Extrapolation with the interpolation, which might help us to understand both. BTW, I am sorry that my mother language is not English, and I am a newbie in Wiki here. I do hope somebody could help me to tidy this article.

First, both the Interpolations and Extrapolations must be consistent with the sampling data. That is, y(x)=yi when x=xi.

Second, We need to give the y(x) for Xi<x<Xi+1 , that is the task of Interpolation. We need to give the y(x) for x<X1 and x>Xn, that is the task of Extrapolation.

Third, Any Interpolation will tend to have one or more corresponding Extrapolation. For example, the nearest-neighbour constant Interpolation will have a corresponding nearest-neighbour constant Extrapolation. Example 2, the linear interpolation will have a corresponding linear extrapolation.

Fourth, when we are using the Extrapolation, it almost always work with a corresponding Interpolation method at the same time. So Extrapolation seems more like an extra attachment of the Interpolation.

Fiveth, We can combine the Extrapolation and the Interpolation together. Nearest-neighbour Extrapolation combined with Linear Interpolation is a good sample. —Preceding unsigned comment added by Nodeed (talk • contribs) 15:00, 30 April 2008 (UTC)

[edit] Removing statistics content

I am removing the sentence ..."The extent to which an extrapolation is accurate is known as the "prediction confidence interval", and is usually expressed as an upper and lower boundary within which the prediction is expected to be accurate 19 times out of 20 (a 95% confidence interval)." ... on the grounds that while this might be possible it requires a large set of context and assumptions which are not stated. Also removing from Category:Statistics for now. Of course there is the possibility of including some properly contextualised statistical content. Melcombe (talk) 15:52, 13 May 2008 (UTC)