Talk:Estimation

This  level-5 vital article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Statistics High‑importance This article is within the scope of WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.StatisticsWikipedia:WikiProject StatisticsTemplate:WikiProject StatisticsStatistics High This article has been rated as High-importance on the importance scale.

Why does this article mention software? Estimation is a much, much broader topic than that. It's a way of using math to cope with the world, isn't it? In industry, estimation probably is most often associated with estimating the construction cost of buildings. Software development seems a very small part of the story of estimation.

There is also estimation in the Harry Van Trees sense, applied to random variables, where we use statistical knowledge to estimate the value of a variable we can't observe based on some other variable that we can observe. There are various special cases of this such as detection, where the hidden variable is a boolean. Estimation is some of the mathematics behind modems and radars. (The hidden variable for the modem is "what bit did the sender think he was sending?" and for the radar "is there an aircraft up there, and what is its precise location?"). The same math is also used in medical imaging.

http://www.amazon.com/exec/obidos/tg/detail/-/0471095176/104-9158830-1284763

http://web.mit.edu/6.432/www/

I took this course a few years back and it was way cool. I know MIT is working on putting their lecture notes online, I'm really looking forward to when this course is there.

This antedates Harry Van Trees! See Lehmann's book Theory of Point Estimation. Ronald Fisher is one of the big names in estimation theory. Michael Hardy 16:14, 18 Dec 2003 (UTC)

"often based on approximate, uncertain, incomplete, or noisy data". I disagree. "based on incomplete data" is better. Bad data does not make the result an estimate - it is just a bad result. Insufficient data makes the result an estimate. See inferential statistics. Bo Jacoby 09:10, 9 September 2005 (UTC)Reply

How are the different estimate scenarios 50/50, 70/30 and 90/10 arrived at? —The preceding unsigned comment was added by 134.146.0.27 (talk) 09:20, 15 May 2007 (UTC).Reply

I have moved the text on estimation in project management to a dedicated article. --Dan Polansky 12:06, 15 October 2007 (UTC)Reply

how to do the project of software company in the details. —Preceding unsigned comment added by 117.242.18.119 (talk) 09:17, 13 October 2009 (UTC)Reply

The approximation is for round numbers, probably the exact number is unknown, but surely the information is enough accurate to be useful, the estimation is just something that you believe of something that is uncertain. For example, you can estimate that you are going to do a certain work in 30 minutes, but probably the work may be easier than what you expected and take less than 20 minutes, or the work may be so hard and you can take more than 40 minutes. --BOMBINI (messages) 04:06, 16 May 2012 (UTC)Reply

I have modified the article into a disambiguation page as there were/are many different interpretations of "esimation" (and "estimate", which redirects here) in different applications. Melcombe (talk) 11:41, 16 May 2012 (UTC)Reply

Unless you can demonstrate that the various kinds of "estimation" addressed do not involve the generation of an "estimate", such a disambiguation page is a clear violation of WP:DABCONCEPT. The primary topic of an article may be abstract and difficult to write about, but that does not eliminate the fact that there is a single primary topic of which all of these kinds of estimation are merely subsets. bd2412 T 19:24, 9 August 2012 (UTC)Reply

A guesstimate is just a special case of an estimate. Since the latter article is short, and not likely to grow substantially in its current location, I propose to merge it here and make it a section of Estimation. bd2412 T 21:46, 5 November 2014 (UTC)Reply

@BD2412: I think it's a separate concept, like back-of-the-envelope calculation is distinct from calculation. I'm more in favor of emptying estimation, moving portions into Statistical estimation and Approximation, then merging the rest with Estimation (disambiguation). Portions of the present article read too much like a textbook. Do you think the style in Wikipedia:Manual of Style/Disambiguation pages is feasible? Fgnievinski (talk) 04:09, 6 November 2014 (UTC)Reply

• Estimation is not an ambiguous topic, it is merely abstract. Emptying it out would leave no place for links to point that intended to discuss the general concept rather than either of those subtopics, and no place for the reader to find the basic overview of the subject. bd2412 T 04:27, 6 November 2014 (UTC)Reply