Hi,
I have recently begun using R in my personal research.
I have a pair of
bivariate probit models that I have already estimated in Stata (v. 8.0)
that I am trying to replicate in R. I have been able to replicate the
first Biprob using R that is run on a Windows based system. However, the
Windows based system does not have enough RAM to run two Biprobs and then
generate an ROC curve comparing the fitted values.
One way to do this in Windows might be
run one bivariate probit model and save the workspace by
save.image("filename.RData")
run another bivariate probit model and save the workspace with a different
name.
load those workspaces by load("filename.RData") command, and run an ROC
curve.
I have been able to access the university's
mainframe through the help of
a professor in my department. However, to run R (version released on
8/25) on the university mainframe I need to use Linux. I have been able
to load the data into the system. However, when I try to run the same
Biprob model(the one I replicated earlier on the Windows system) it runs
the VGAM prompts and then I get the following error command:
Error: incorrect number of subscripts on matrix
That's a little strange. Could you give us what you get by
typing traceback() after your error? Also, what is the syntax you used to
run the model?
Kosuke
---------------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601
Department of Politics eFax: 973-556-1929
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012
http://www.princeton.edu/~kimai
---------------------------------------------------------
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options:
http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/