If the dependent variable is not included in treat1 (which is not
likely if you've used setx to create it), then you have to exclude it
also from the original data. You can do this by specifying the
original dataset as data and then adding the regression equation
without the dependent variable:
w <- whatif(formula = ~ x1 + x2 + x3, data = yourdata, cfact = treat1)
fabrizio
On 1 May 2008, at 13:18, Keith Schnakenberg wrote:
When I give WhatIf a command like this:
summary(whatif(data= model1, cfact= treat1))
where model1 is the zelig model and treat1 is a setx object, I get
the following error:
Error in whatif(data = model1, cfact = treat1) :
the list supplied to 'data' is not a valid output object
I am not sure why this is the case. I did not get this error using
the same setup on last week's problem set, for example. What might be
some reasons for this?
_______________________________________________
gov2001-l mailing list
gov2001-l at
lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
---------------------------------------------------------
Fabrizio Gilardi
Weatherhead Center for International Affairs
Harvard University
1737 Cambridge Street
Cambridge, MA 02138 (USA)
Phone +1 617 495 1467
Fax +1 617 495 8292
fgilardi at
wcfia.harvard.edu
www.fabriziogilardi.org
---------------------------------------------------------