Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Standard error of the estimate for svy: reg


From   Steven Samuels <sjhsamuels@earthlink.net>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Standard error of the estimate for svy: reg
Date   Wed, 22 Aug 2007 23:39:31 -0400

One More line, one 'last time'

Mike-I apologize. I certainly accept "Standard Error of Regression" and "Standard Error of the Estimate" as valid synonyms for for the residual S.D. I offered the citation to indicate why I was surprised at the usage "Standard Error of the Estimate"; my training had taught a distinction between a standard error (standard deviation of an estimate, a quantity which depends on sample size) and standard deviation, a fixed characteristic of a population.

In fact, I see the logic of the term "Standard Error of Regression". The model quantities (observation - mean) are commonly referred to both as "deviations" from and as "error terms" or "errors"; hence I can easily understand that "standard deviation of regression" and "standard error of regression" are synonyms. The logic of the term "Standard Error of the Estimate" escapes me--I wonder where it originated--but I can certainly accept it. When I taught introductory statistics, I tried to alert the students to all the terms they would encounter for "Mean Square Error" and "Residual S.D." I should not have been surprised to find two more. Thanks to you and to Richard to enlarging my vocabulary.
,

Steve




sjhsamuels@earthlink.net
18 Cantine's Island
Saugerties, NY 12477
Phone: 845-246-0774
EFax: 208-498-7441




On Aug 22, 2007, at 4:31 PM, Michael Hanson wrote:


On Aug 22, 2007, at 3:23 PM, Steven Samuels wrote:

I'll leave this topic with the following reference:

http://www.bmj.com/cgi/content/full/331/7521/903

Steve
Since you've left this topic, you've foreclosed any opportunity to clarify the relevance of that citation. In my very quick read of it, nothing in the second- and third-to-last paragraphs seems to be inconsistent with the use of the terminology "standard error of the estimate" (or "standard error of the regression") that we have previously established is not uncommon in certain social sciences but apparently unknown to at least some people in other (biomedical?) fields; the remaining paragraphs appear to discuss other topics. I personally don't see a problem with different fields having different nomenclatures, and the point of my previous message was simply to indicate that the questioned term is not uncommon in certain (broad) fields of applied statistics.

With the discussion abruptly ended I gather that the implication was meant to be that certain fields (viz. the social sciences) were misusing a term. I would be interested in learning the substance of that argument, if in fact that was the case. Otherwise I am still puzzling over the contribution of the citation to the prior exchange.

-- Mike

P.S.: Stas's argument against the use of the RMSE/SEE in the OP's question sounds valid to me, but I do not claim any familiarity with estimation using survey data.

*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
Steven  Samuels

sjhsamuels@earthlink.net
18 Cantine's Island
Saugerties, NY 12477
Phone: 845-246-0774
EFax: 208-498-7441




*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index