Hello
After (simple random cluster) resampling with replacement I ran MSM
function and I'm getting the following warning message ,which I'm not
sure why. I don't have any absorbing stage set up in my MSM model. I
have a 4 stage uni-directional MSM model. The only thing I see might be
a problem is that when the last stage (stage 4 in my data) gets repeated
it seems to give me a warning message. although, it picks only one
observation whereas there might be other such cases in the data
(printing only the first case?)
> out1.msm2
<-msm(stage~years,subject=patient,data=mydata2,qmatrix=new.twoway9.q,con
trol=list(trace=1,report=1),covariates=~black+tport11,constraint=list(bl
ack=c(1,1,1),tport11=c(1,1,1)))
...
Warning message:
Absorbing - absorbing transition at observation 30 in:
msm.check.model(msmdata$fromstate, msmdata$tostate, msmdata$obs,
My data at observation around 30 look like ('years' is the time
variable)
> mydata2[27:35,]
count patient black years stage tport11 pcount
1 1029 1 0 1 1 17
1 1029 1 0 1 1 17
2 1029 1 19 4 1 17
2 1029 1 19 4 1 17
1 1033 0 0 1 1 20
1 1033 0 0 1 1 20
2 1033 0 18 4 1 20
2 1033 0 18 4 1 20
I'm wondering if I can ignore this warning message or should I have not
done resampling with replacement. Any input would be greatly
appreciated.
Thank you so much for your time,
kelly
-------------------------------------------------------------
KyungAh (Kelly) Im
Epidemiology Data Center
Graduate School of Public Health
Department of Epidemiology
University of Pittsburgh
PA 15261
PHONE:(412) 624- 4612 FAX: (412) 624- 3775
Email: Im at edc.pitt.edu
http://www.edc.gsph.pitt.edu
