Bruno Estigarribia
2009-Jan-22 05:32 UTC
[R] Standard errors of least squares adjusted means
Hello, I have the following model: lm.7 <- lm(Y ~ F + C1 + C2 , data = EM4) F is a 4-level factor, the rest are covariates centered at their mean (Y is a two-column matrix). I have tried to find functions to give the model-adjusted means (adjusted at the covariates'means) and their standard deviations for each. (That is, what I believe is called in SAS "least square or LS-means, whose errors one obtains by STDERR) I have tried help.search and RSiteSearch with several terms including "standard errors", "least square means", "adjusted means". I have found how to extract the SE from coefficients (se.coef from package arm), or the SE for contrasts (se.contrast from package stats), but not the SE for an adjusted mean. Thank you, -- Bruno Estigarribia Postdoctoral Fellow FPG Child Development Institute Neurodevelopmental Disorders Research Center University of North Carolina at Chapel Hill 105 Smith Level Rd Chapel Hill, NC 27599-8180 USA + 1 (919) 843-7685
Hi Bruno, Apropos of "ls-means" ...>> I have tried help.search and RSiteSearch with several terms including >> "standard errors", "least square means", "adjusted means".And "ls-means," which is what "you" call them? There are many threads on this, spanning many years. The following, RSiteSearch("ls-means") will get you to some, e.g.: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/103520.html Read through to the bottom. There are much better ways of examining/summarizing, your model. See, for instance, the effects package of Prof. Fox and functions in the Design package of Prof. Harrell. HTH, Mark. Bruno Estigarribia wrote:> > Hello, > > I have the following model: > lm.7 <- lm(Y ~ F + C1 + C2 , data = EM4) > F is a 4-level factor, the rest are covariates centered at their mean (Y > is a two-column matrix). > I have tried to find functions to give the model-adjusted means > (adjusted at the covariates'means) and their standard deviations for each. > (That is, what I believe is called in SAS "least square or LS-means, > whose errors one obtains by STDERR) > I have tried help.search and RSiteSearch with several terms including > "standard errors", "least square means", "adjusted means". I have found > how to extract the SE from coefficients (se.coef from package arm), or > the SE for contrasts (se.contrast from package stats), but not the SE > for an adjusted mean. > Thank you, > > -- > Bruno Estigarribia > Postdoctoral Fellow > FPG Child Development Institute > Neurodevelopmental Disorders Research Center > University of North Carolina at Chapel Hill > > 105 Smith Level Rd > Chapel Hill, NC 27599-8180 > USA > + 1 (919) 843-7685 > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > >-- View this message in context: http://www.nabble.com/Standard-errors-of-least-squares-adjusted-means-tp21598897p21599542.html Sent from the R help mailing list archive at Nabble.com.
Dear Bruno, See the effects package (on CRAN) for a generalization of adjusted means. I hope this helps, John ------------------------------ John Fox, Professor Department of Sociology McMaster University Hamilton, Ontario, Canada web: socserv.mcmaster.ca/jfox> -----Original Message----- > From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]On> Behalf Of Bruno Estigarribia > Sent: January-22-09 12:32 AM > To: r-help at r-project.org > Subject: [R] Standard errors of least squares adjusted means > > Hello, > > I have the following model: > lm.7 <- lm(Y ~ F + C1 + C2 , data = EM4) > F is a 4-level factor, the rest are covariates centered at their mean (Y > is a two-column matrix). > I have tried to find functions to give the model-adjusted means > (adjusted at the covariates'means) and their standard deviations for each. > (That is, what I believe is called in SAS "least square or LS-means, > whose errors one obtains by STDERR) > I have tried help.search and RSiteSearch with several terms including > "standard errors", "least square means", "adjusted means". I have found > how to extract the SE from coefficients (se.coef from package arm), or > the SE for contrasts (se.contrast from package stats), but not the SE > for an adjusted mean. > Thank you, > > -- > Bruno Estigarribia > Postdoctoral Fellow > FPG Child Development Institute > Neurodevelopmental Disorders Research Center > University of North Carolina at Chapel Hill > > 105 Smith Level Rd > Chapel Hill, NC 27599-8180 > USA > + 1 (919) 843-7685 > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html> and provide commented, minimal, self-contained, reproducible code.