Hi there!
I'm trying to use the R package MSBVAR to generate a conditional forecast from
a bayesian VAR. The condition that I am imposing for the forecast is observed
values for all but one of the variables. I then want to forecast the path of
that one variable, given observed values for all others.
I have no problem importing the data and estimating the VAR, however, when I
try and use the command "hc.forecast" I get the following errors:
Error in solve.default((hstar.update), (crossprod(X.update, Y.update) + :
system is computationally singular: reciprocal condition number = 3.6882e-17
Error in chol.default(vcv.Bh) :
the leading minor of order 88 is not positive definite
This is how I am using the command:
conditional<-hc.forecast(flatvar,condition,20,burnin=3000,gibbs=5000)
where flat var is a bayesian VAR with a flat prior, condition is a matrix that
specifies the condition mentioned above (observed values, and 0's for the
variable I want to forecast), 20 is the number of steps I want to forecast.
Any help is very much appreciated. I'm new to this so if I am leaving out
information please let me know!
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
