[R] Ex ante forecasting from structural equation models (SEM package)

John Fox jfox at mcmaster.ca
Tue Aug 13 16:26:08 CEST 2002

Dear Bernhard,

What you want are the reduced-form coefficients implied by the structural 
form of the model. I don't know a simple way of getting this from the RAM 
formulation of the model, short of constructing the coefficient matrices A 
and B from the estimates, which shouldn't be too onerous.

As far as I know, there are no methods for objects of class "sem" other 
than what's included in the package. Suggestions are welcome, of course.


At 01:09 PM 8/13/2002 +0200, Pfaff, Bernhard wrote:
>Dear Helplist,
>I want to produce forecasts from a structural equation model. With the SEM
>package the model setup and its estimation is possible. However, I have not
>figured out how to obtain ex ante forecasts, i.e. applying the Gauss-Seidel
>algorithm to the estimated structural equations for provided values of the
>exogenous variables (i.e.: y_t = -inv(A)*B*x_t).
>Does anyone know if the there is a function for producing such forecasts via
>Gauss-Seidel or what other functions are supporting objects of class 'SEM'
>or model notations in 'RAM' for achieving this task?
>Many thks,

John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: jfox at mcmaster.ca
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox

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