Dear all,
I'm trying to fit 2SLS with systemfit(). Unfortunately, it
somehow doesn't allow me to add any pre-treatment covars to
the first and second stages. Without any covars (i.e.
without p055.k) it works just fine and also produces the
correct result. Does anybody know what's wrong here? (My
vars are not perfectly collinear; I've checked that of course.)
Thanks,
Holger
first <- papilla.intake.N99 ~ contba.m + p055.k
second <- weight.99 ~ papilla.intake.N99 + p055.k
inst <- ~ contba.m
system <- list(demand=first, supply=second)
out.1 <- systemfit("2SLS", system,
inst=inst, data=data.imp1,
+ maxiter=1000)
Error in solve.default(crossprod(Xf), crossprod(Xf, Y), tol
= solvetol) :
system is computationally singular: reciprocal
condition number = 5.27922e-018
--
Holger Lutz Kern
Graduate Student
Department of Government
Cornell University
Institute for Quantitative Social Science
Harvard University
1737 Cambridge Street N350
Cambridge, MA 02138
www.people.cornell.edu/pages/hlk23