Comments on: R examine objects tutorial http://www.win-vector.com/blog/2009/11/r-examine-objects-tutorial/?utm_source=rss&utm_medium=rss&utm_campaign=r-examine-objects-tutorial The Applied Theorist's Point of View Thu, 15 Jul 2010 00:13:50 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: jmount http://www.win-vector.com/blog/2009/11/r-examine-objects-tutorial/comment-page-1/#comment-1643 jmount Wed, 09 Dec 2009 16:21:55 +0000 http://www.win-vector.com/blog/?p=1134#comment-1643 <a href="#comment-1613" rel="nofollow">@Dirk Eddelbuettel </a> Unfortunately coef(model) does not return the standard errors. coef() is also hard to find (not listed in methods(class=class(model)), and being a generic functions you can't find it unless you notice its link in the documentation). This is also a case where the documentation is irritating. The entire text in the help(lm) page is: "See also: The generic functions coef, effects, residuals, fitted, vcov." You are not told why to look at these functions, so you have to go by their names. And coef() is a generic function so its documentation is the nearly meaningless phrase "coef is generic functions which extracts model coefficients" (the problem being what is extracted will vary from model type to model type). @Dirk Eddelbuettel Unfortunately coef(model) does not return the standard errors. coef() is also hard to find (not listed in methods(class=class(model)), and being a generic functions you can’t find it unless you notice its link in the documentation).

This is also a case where the documentation is irritating. The entire text in the help(lm) page is: “See also: The generic functions coef, effects, residuals, fitted, vcov.” You are not told why to look at these functions, so you have to go by their names. And coef() is a generic function so its documentation is the nearly meaningless phrase “coef is generic functions which extracts model coefficients” (the problem being what is extracted will vary from model type to model type).

]]> By: Dirk Eddelbuettel http://www.win-vector.com/blog/2009/11/r-examine-objects-tutorial/comment-page-1/#comment-1613 Dirk Eddelbuettel Tue, 01 Dec 2009 20:34:02 +0000 http://www.win-vector.com/blog/?p=1134#comment-1613 Very nice indeed -- but the form of using 'coef(model)' is generally preferred over 'summary(model)$coefficients'. Very nice indeed — but the form of using ‘coef(model)’ is generally preferred over ‘summary(model)$coefficients’.

]]> By: shane http://www.win-vector.com/blog/2009/11/r-examine-objects-tutorial/comment-page-1/#comment-1600 shane Thu, 26 Nov 2009 04:05:35 +0000 http://www.win-vector.com/blog/?p=1134#comment-1600 very helpful, thanks! very helpful, thanks!

]]> By: jmount http://www.win-vector.com/blog/2009/11/r-examine-objects-tutorial/comment-page-1/#comment-1586 jmount Sun, 22 Nov 2009 01:38:58 +0000 http://www.win-vector.com/blog/?p=1134#comment-1586 A sincere thank you to Joshua Ulrich ( http://www.linkedin.com/in/joshuaulrich ) who pointed out in this case the answers are in the R documentation system. I guess I had been getting too cynical to thoroughly look there. A sincere thank you to Joshua Ulrich ( http://www.linkedin.com/in/joshuaulrich ) who pointed out in this case the answers are in the R documentation system. I guess I had been getting too cynical to thoroughly look there.

]]>