Question

我尝试使用我自己的数据集运行R文档中提供的针对nlmeODE包的Theoph模型，并且能够返回参数估计值。但是，当我添加一个我想估计的附加参数时，我收到了一条错误消息，如下所示。谁能告诉我哪里出了问题以及如何解决这个问题呢？

R文档中提供的示例模型如下所示：

data(Theoph)
TheophODE <- Theoph
TheophODE$Dose[TheophODE$Time!=0] <- 0
TheophODE$Cmt <- rep(1,dim(TheophODE)[1])
OneComp <- list(DiffEq=list(
dy1dt = ~ -ka*y1 ,
dy2dt = ~ ka*y1-ke*y2),
ObsEq=list(
c1 = ~ 0,
c2 = ~ y2/CL*ke),
Parms=c("ka","ke","CL"),
States=c("y1","y2"),
Init=list(0,0))
TheophModel <- nlmeODE(OneComp,TheophODE)

Theoph.nlme <- nlme(conc ~ TheophModel(ka,ke,CL,Time,Subject),
data = TheophODE, fixed=ka+ke+CL~1, random = pdDiag(ka+CL~1),
start=c(ka=0.5,ke=-2.5,CL=-3.2),
control=list(returnObject=TRUE,msVerbose=TRUE),
verbose=TRUE)
plot(augPred(Theoph.nlme,level=0:1))

我的修改后的模型如下：

group1 <- groupedData(conc ~ Time | Subject,
                  data=myowndata)

OneComp2 <- list(DiffEq=list(
  dy1dt = ~ -ka*y1 + kr*y2,
  dy2dt = ~ ka*y1 - kr*y2 -ke*y2), #I added in a reabsorption rate kr
  ObsEq=list(
    c1 = ~ 0,
    c2 = ~ y2/CL*ke),
  Parms=c("ka","ke","kr","CL"),
  States=c("y1","y2"),
  Init=list(0,0))

Model2 <- nlmeODE(OneComp2,group1)

nlme2 <- nlme(conc ~ Model2(ka,ke,kr,CL,Time,Subject),
                  data = group1, fixed=ka+ke+kr+CL~1, random = pdDiag(ka+CL~1),
                  start=c(ka=0.5,ke=-2.5,kr=-2.5,CL=-3.2),
                  control=list(returnObject=TRUE,msVerbose=TRUE),
                  verbose=TRUE)

错误消息显示为：

Error in numericDeriv(form[[3L]], names(ind), env) : 
  Missing value or an infinity produced when evaluating the model

当nlmeODE的R文档中的Theoph模型示例中添加了其他参数时，nlmeODE不起作用

0 个答案: