Dear Xionqing,
It's still difficult to say without the data, but I guess the error message
reads "System computation is singular", meaning you just have too many
variables or some variables are near-aliases of eachother. I did not ask to
read the posting guide to be difficult. Without some reproducible code and
detailed information on the data, it's impossible to say what is the problem
exactly.
Cheers
Joris
On Wed, Jun 9, 2010 at 2:47 AM, Xiongqing Zhang <xqzhang85@yahoo.com>
wrote:
> Dear Joris,
>
> Thank for your reply.
>
>
> Firstly, I introduced one dummy variable (eg: u=0, the first group; u=1,
> the second group) to the model, and I got the results (parameter estimators
> and AIC value).
>
> #zero inflated negative binomial model
>
> library(pscl)
>
> dat=read.csv("data.csv",header=T)
>
> zinb <- zeroinfl(y ~x1+x2+x3+x4+x5+u | x1+x2+x3+x4+x5+u, data = dat,
link> "logit", dist = "negbin")
>
> summary(zinb)
>
>
>
> But the results are not satisfied to me. So I decided to introduce two
> dummy variables (u1,u2) to the model, and then I could not get the results.
>
>
>
> dat=read.csv("data.csv",header=T)
>
> zinb <- zeroinfl(y ~x1+x2+x3+x4+x5+u1+u2 | x1+x2+x3+x4+x5+u1+u2, data
> dat, link= "logit", dist = "negbin")
>
> summary(zinb)
>
> And the error message is
>
> “error in solve.default(as.matrix(fit$hessian)) :
>
> System computation is unsigular
>
> moreover: Warning message:
>
> In glm.fit(Z, as.integer(Y0), weights = weights, family = binomial(link
> linkstr), :
>
> algorithm did not converge”
>
>
>
> Again, many thanks for you.
>
>
>
> Sincerely,
>
> Xiongqing
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> --- On *Tue, 6/8/10, Joris Meys <jorismeys@gmail.com>* wrote:
>
>
> From: Joris Meys <jorismeys@gmail.com>
> Subject: Re: [R] Please help me
> To: "Xiongqing Zhang" <xqzhang85@yahoo.com>
> Cc: r-help@r-project.org
> Date: Tuesday, June 8, 2010, 11:22 AM
>
>
> First, read the posting guides.
> Then, supply us with a bit more information, like the package you
> used, example code that reproduces the error, information about the
> data, the complete error message, the traceback (use the function
> traceback() right after you got the error).
>
> Otherwise we ain't going to be able to help you.
>
> Cheers
> Joris
>
> On Tue, Jun 8, 2010 at 9:42 AM, Xiongqing Zhang
<xqzhang85@yahoo.com<http://cn.mc453.mail.yahoo.com/mc/compose?to=xqzhang85@yahoo.com>>
> wrote:
> >
> > Dear Mr. or Ms.,
> >
> > I used the R-software to run the zero-inflatoin negative binomial
model
> (zeroinfl()) .
> >
> > Firstly, I introduced one dummy variable to the model as an
independent
> variable, and I got the estimators of parameters. But the results are not
> satisfied to me. So I introduced three dummy variables to the model. but I
> could not get the results. And the error message is
> "solve.default(as.matrix(fit$gaussian)) ". I do not know the
reasons fully.
> And I want to know what method was used to estimate the parameters when I
> called the function zeroinfl() .
> >
> > I will be very appreciate if you can help me. Thank you.
> >
> > Best regards,
> >
> > Sincerely,
> > Xiongqing Zhang
> >
> >
> >
> > [[alternative HTML version deleted]]
> >
> >
> > ______________________________________________
> >
R-help@r-project.org<http://cn.mc453.mail.yahoo.com/mc/compose?to=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<http://www.r-project.org/posting-guide.html>
> > and provide commented, minimal, self-contained, reproducible code.
> >
> >
>
>
>
> --
> Joris Meys
> Statistical consultant
>
> Ghent University
> Faculty of Bioscience Engineering
> Department of Applied mathematics, biometrics and process control
>
> tel : +32 9 264 59 87
>
Joris.Meys@Ugent.be<http://cn.mc453.mail.yahoo.com/mc/compose?to=Joris.Meys@Ugent.be>
> -------------------------------
> Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php
>
>
>
--
Joris Meys
Statistical consultant
Ghent University
Faculty of Bioscience Engineering
Department of Applied mathematics, biometrics and process control
tel : +32 9 264 59 87
Joris.Meys@Ugent.be
-------------------------------
Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php
[[alternative HTML version deleted]]