Hi,
I am using flexmix to model some data which is modelled with linear
regressions. I have results obtained along the lines of that shown below, and
can retreive component parameters, but I cannot find a way or retrieving the
variance of the sigma (variance of the normal model) can anyone advise how to do
this?:
>fm=compiledstepFlexmix(lkopsum ~
lfasttax+ypr+ypr2,data=land,k=k,nrep=2000,control=Fcontrol)
>fmexample=fm@models$'3'
fmexample
Call:
compiledstepFlexmix(lkopsum ~ lfasttax + ypr + ypr2, data = land,
control = Fcontrol, k = 3, nrep = 500)
Cluster sizes:
1 2 3
325 425 216
convergence after 57 iterations> refitexample=refit(fmexample)
> summary(refitexample)
$Comp.1
Estimate Std. Error z value Pr(>|z|)
(Intercept) 8.3713863 0.1165509 71.8260 < 2.2e-16 ***
lfasttax 0.7921171 0.0190065 41.6762 < 2.2e-16 ***
ypr 0.0641430 0.0048686 13.1747 < 2.2e-16 ***
ypr2 0.0044674 0.0012131 3.6827 0.0002308 ***
---
Signif. codes: 0 â***â 0.001 â**â 0.01 â*â 0.05 â.â 0.1 â â 1
$Comp.2
Estimate Std. Error z value Pr(>|z|)
(Intercept) 6.2262676 0.1340461 46.449 <2e-16 ***
lfasttax 1.0864937 0.0196605 55.263 <2e-16 ***
ypr 0.1129625 0.0057275 19.723 <2e-16 ***
ypr2 -0.0029301 0.0010706 -2.737 0.0062 **
---
Signif. codes: 0 â***â 0.001 â**â 0.01 â*â 0.05 â.â 0.1 â â 1
$Comp.3
Estimate Std. Error z value Pr(>|z|)
(Intercept) 8.4120331 0.3045145 27.6244 < 2.2e-16 ***
lfasttax 0.6394828 0.0504086 12.6860 < 2.2e-16 ***
ypr 0.0701468 0.0155021 4.5250 6.04e-06 ***
ypr2 0.0110253 0.0034031 3.2397 0.001196 **
---
Signif. codes: 0 â***â 0.001 â**â 0.01 â*â 0.05 â.â 0.1 â â 1
[[alternative HTML version deleted]]