[R] Re: S+ code

Peter Malewski malewski.peter at MH-Hannover.de
Mon Jul 12 14:42:23 CEST 1999

Dear Mr. Cleland, SEM-ueser, R-User, ...

    there isn't s-plus or R (a GNU-free version of the programming language
S) code for structural equation modelling. I always wondered why. I
personally think the language S, with the implementations S+ and R, is a
good-perhaps the best-environment for statistical analyses. Why? It is
object orientated, logical, etc. For example: the code for least squares
regression is: lm(y~x). And isn’t it a waste of time to learn for a new
analyses a new programming language?

Several month ago I thought about including structural eq. modelling  in
R---either by writing an interface to the MX-Language (which unfortunately
isn’t under the GNU-licence but nearly “free”) or built it up from scratch.
I haven’t because I haven’t got time and ...

Surely I the philosophies of the people who use S and SEM-Models are
different. Personally I think that this is the main reason why there isn’t
code for SEM for S. Perhaps one of the R-users might explain why there SEM
should not be included in R...

Greetings from an R but not yet SEM user (perhaps never)


ps. Sorry for the bad Engish...

Chuck Cleland wrote:

> Boechler, Mike wrote:
> >
> > Good day,
> > Is anyone aware of any objects written in S+ that allow one to fit
> > structural equation models (latent variable and otherwise)? Thanks in
> > advance.
> > Mike
>   I am not aware of any libraries or functions, but you might try
> posting the question on the S-Plus list.
> s-news at wubios.wustl.edu
>   Also check the S-Plus archives for available libraries and functions.
> http://lib.stat.cmu.edu/S/
> AND Windows
> http://lib.stat.cmu.edu/DOS/S/
> Chuck Cleland


* P. Malewski
* Hannover Medical School
* Department of Psychosomatics and Psychotherapy
* Carl-Neuberg-Str. 1
* Tel.:+49 511 532 3194
* Fax.:+49 511 532 3190

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list