Re: [R] Control variables in mediation analysis
Well, yeah ... but it doesn't necessarily follow that the result converges to what you want. Bootstrapping extreme order statistics is an example. -- Bert On Thu, Mar 21, 2013 at 1:49 PM, Ista Zahn wrote: > Hi, > > Just about anything can be bootstrapped, that's one of the things that > makes bootstrapping great! Use e.g., the boot package, and bootstrap > the product of the coefficients. > > Alternatively the mediation package > (http://cran.r-project.org/web/packages/mediation/index.html) might > help, though I'm not sure if it uses bootstrap. > > Best, > Ista > > On Thu, Mar 21, 2013 at 3:21 PM, F.Doerwald > wrote: > > Hi everyone,I would like to test a mediation model, that has several > control variables. More specifically, I would like to test the indirect > effect with bootstrapping. However, all the packages I have found so far > (e.g. MBESS) only allow testing a simple mediation model (One independent, > one mediator, one dependent) so that I cannot include any controls. > > Can somebody help me? > > > > > > [[alternative HTML version deleted]] > > > > __ > > R-help@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. > > __ > R-help@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. > -- Bert Gunter Genentech Nonclinical Biostatistics Internal Contact Info: Phone: 467-7374 Website: http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm [[alternative HTML version deleted]] __ R-help@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.
Re: [R] Control variables in mediation analysis
Dear Ista and F.Doerwald, F.Doerwald could also fit a model using sem() in the sem package, and then use deltaMethod() in the car package to get the estimated value and standard error of any function of model coefficients that he or she chooses. Sorry, I didn't read the original posting. I hope this helps, John John Fox Sen. William McMaster Prof. of Social Statistics Department of Sociology McMaster University Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox/ On Thu, 21 Mar 2013 16:49:46 -0400 Ista Zahn wrote: > Hi, > > Just about anything can be bootstrapped, that's one of the things that > makes bootstrapping great! Use e.g., the boot package, and bootstrap > the product of the coefficients. > > Alternatively the mediation package > (http://cran.r-project.org/web/packages/mediation/index.html) might > help, though I'm not sure if it uses bootstrap. > > Best, > Ista > > On Thu, Mar 21, 2013 at 3:21 PM, F.Doerwald wrote: > > Hi everyone,I would like to test a mediation model, that has several > > control variables. More specifically, I would like to test the indirect > > effect with bootstrapping. However, all the packages I have found so far > > (e.g. MBESS) only allow testing a simple mediation model (One independent, > > one mediator, one dependent) so that I cannot include any controls. > > Can somebody help me? > > > > > > [[alternative HTML version deleted]] > > > > __ > > R-help@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. > > __ > R-help@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. __ R-help@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.
Re: [R] Control variables in mediation analysis
Hi, Just about anything can be bootstrapped, that's one of the things that makes bootstrapping great! Use e.g., the boot package, and bootstrap the product of the coefficients. Alternatively the mediation package (http://cran.r-project.org/web/packages/mediation/index.html) might help, though I'm not sure if it uses bootstrap. Best, Ista On Thu, Mar 21, 2013 at 3:21 PM, F.Doerwald wrote: > Hi everyone,I would like to test a mediation model, that has several control > variables. More specifically, I would like to test the indirect effect with > bootstrapping. However, all the packages I have found so far (e.g. MBESS) > only allow testing a simple mediation model (One independent, one mediator, > one dependent) so that I cannot include any controls. > Can somebody help me? > > > [[alternative HTML version deleted]] > > __ > R-help@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. __ R-help@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.
