I tried to post this question two times, each time it seemed to fail, since
"
An embedded and charset-unspecified text was scrubbed..."
So I try it again:
I fitted a mixture density of two gaussians two my data. I now want to
caluclate the standard errors of the estimates via the boot.se command of
the mixtools package. My question is now, if the output is correct? It
seems a bit odd to me, so is this correct what I am doing and can I rely on
the values?
My data can be found at:
http://s000.tinyupload.com/?file_id=09285782882980618119
my code:
normalmix<-normalmixEM(dat,k=2,lambda=c(0.99024,(1-0.99024)),fast=FALSE,maxit=10000,epsilon=1e-16,maxrestarts=1000)
normalmix$loglik
normalmix$lambda
se<-boot.se(normalmix,B=1000)
se$lambda.se
se$mu.se
se$sigma.se
final results:
labmdahat=0.99024
mu1hat=-0.00115
mu2hat=0.040177
sigma1hat=0.01222
sigma2hat=0.00325
My problem is - and thats why I feel uncomfortable about relying on the
values- that the ouput of boot.se(normalmix) varies quite strong. So
without chanigng the code an drerun it (with the same normalmix, so
normalmix is not rerun again) I get different esimates of the standard
errors. I increased the default value for B from 100 to 1000. In the manual
there is nothing said about any other randomness. So where does it com
from? What should I do now?
[[alternative HTML version deleted]]
