[Mediation-information] Continuous treatments in mediation package

dustin tingley dtingley at gov.harvard.edu
Thu Jul 7 17:22:25 CEST 2011


Dear Weichao-

Our apologies for a delay in answering your question.

You can absolutely use continuous treatments. In the latest release of the
package, version 3.0 (to automatically load this library use
install.packages("mediation") using R 2.13, the latest version of R, or
higher in the future).

In 3.0 this is done through treat.value and control.value arguments. More
details are available here:
http://cran.r-project.org/web/packages/mediation/vignettes/mediation.pdf

We've setup a mailing list that you should join so you can stay abreast of
changes and updates to the package. This is available here
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/mediation-information

Note that sensitivity analyses are only worked out for binary 0/1
treatments.

best,
Dustin

Dustin Tingley
Government Department
Harvard University
http://scholar.harvard.edu/dtingley



On Mon, Jul 4, 2011 at 4:39 PM, Kosuke Imai <kimai at princeton.edu> wrote:

> What's the answer to this FAQ?  Maybe, one of you can answer the question
> and cc to the mediation email list?
>
> Thanks,
> Kosuke
>
> Begin forwarded message:
>
> > From: "BAO, WEICHAO" <baow at email.sc.edu>
> > Date: July 5, 2011 2:32:01 AM GMT+09:00
> > To: "kimai at princeton.edu" <kimai at Princeton.EDU>
> > Subject: mediation R Package
> >
> > Dear Dr. Imai,
> >
> > I am trying to use your R package to do a mediation analyis. I have a
> question about the variable treatment(i.e. exposure ). I know the variable
> treatment in your example is discrete. Can the package still be used if the
> treatment variable is continuous (such as the exposure of some metals)?
> >
> >
> > Thanks you very much.
> >
> > Sincerely,
> >
> > Weichao
> >
> >
> > Weichao Bao
> > Aronal School of Public Health
> > University of South Carolina
> > 803-445-6087
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/mediation-information/attachments/20110707/e5bd4356/attachment.htm>


More information about the Mediation-information mailing list