利用r语言msm包跑数据的时候提示numerical overflow in calculating likelihood，该怎么解决呢？

Question

数据有10000多人，纵向随访数据，保留两次以上的随访对象，

library(msm)
library(readxl)
w<-read_excel("C:\Users\HEART\Desktop\2组分合在一起 - 把葡萄糖异常删除 - 副本.xlsx")
statetable.msm(state, id, data=w)

Q<-rbind(c(0,0.076,0.02,0.057,0.018,0.063,0,0),c(0.30,0,0,0,0,0,0.158,0),c(0.095,0,0,0,0,0,0.30,0),c(0.278,0,0,0,0,0,0.174,0),c(0.278,0,0,0,0,0,0.225,0),c(0.266,0,0,0,0,0,0.149,0),c(0,0.056,0.07,0.042,0.015,0.056,0,0.239),c(0,0,0,0,0,0,0.183,0))
w.msm <-msm(state~years,subject=id,data=w,qmatrix=Q,method="BFGS",control=list(fnscale=4000,maxit = 10000))
w.msm

wage.msm<-msm(state~years,subject=id,data=w,qmatrix=Q,covariates=~age,method="BFGS",control=list(fnscale=4000,maxit = 10000))

Error in Ccall.msm(params, do.what = "lik", ...) :
numerical overflow in calculating likelihood

年龄作为协变量时会提示数据溢出，该怎么改正这个错误呢？

Answer 1

https://bbs.pinggu.org/forum.php?mod=forumdisplay&fid=69&digest=1&typeid=1165

Answer 2

请问，楼主你的Q值如何确定的，是个初学者，请教一下：Q<-rbind(c(0,0.076,0.02,0.057,0.018,0.063,0,0),c(0.30,0,0,0,0,0,0.158,0),c(0.095,0,0,0,0,0,0.30,0),c(0.278,0,0,0,0,0,0.174,0),c(0.278,0,0,0,0,0,0.225,0),c(0.266,0,0,0,0,0,0.149,0),c(0,0.056,0.07,0.042,0.015,0.056,0,0.239),c(0,0,0,0,0,0,0.183,0))

Answer 3

请问，楼主你的Q值如何确定的，是个初学者，请教一下：Q<-rbind(c(0,0.076,0.02,0.057,0.018,0.063,0,0),c(0.30,0,0,0,0,0,0.158,0),c(0.095,0,0,0,0,0,0.30,0),c(0.278,0,0,0,0,0,0.174,0),c(0.278,0,0,0,0,0,0.225,0),c(0.266,0,0,0,0,0,0.149,0),c(0,0.056,0.07,0.042,0.015,0.056,0,0.239),c(0,0,0,0,0,0,0.183,0))

Answer 4

请问楼主解决了吗