[R] Scatterplot and Causality
joerg.luedicke at gmail.com
Mon Apr 22 22:55:48 CEST 2013
Nevertheless, some people try:
On Mon, Apr 22, 2013 at 4:31 PM, Bert Gunter <gunter.berton at gene.com> wrote:
> On Mon, Apr 22, 2013 at 1:15 PM, Lorenzo Isella
> <lorenzo.isella at gmail.com> wrote:
>> On Mon, 22 Apr 2013 17:27:05 +0200, Robert Baer <rbaer at atsu.edu> wrote:
>>> It strikes me that this is not a particularly productive approach to
>>> causality, particularly in an observational setting. You would need to
>>> design an experiment where you had a known manipulation of an explanatory
>>> variable and studied the change in a response variable, and then, you came
>>> back with the roles reversed. I don't think R or indeed any statistical
>>> package can help you here.
>> I certainly agree, but here we are not planning an experiment: we have some
>> data sets that were collected having something totally different in mind and
>> we wonder if we can extract any info about causality from them.
> As others have said:
> (Short answer) No.
> (Long Answer) Don't be ridiculous.
> -- Bert
>> R-help at r-project.org mailing list
>> 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
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help