Re: [R] survreg with measurement uncertainties

ents: > model > 2.606 > > > -Original Message----- > From: Kyle Penner [mailto:kpen...@as.arizona.edu] > Sent: Wednesday, June 12, 2013 3:49 PM > To: Terry Therneau > Cc: r-help@r-project.org > Subject: Re: [R] survreg with measurement uncertainties > > Hi Terry

Re: [R] survreg with measurement uncertainties

: r-help@r-project.org Subject: Re: [R] survreg with measurement uncertainties Hi Terry, Thanks for your quick reply. I am talking about uncertainty in the response. I have 2 follow up questions: 1) my understanding from the documentation is that 'id' in cluster(id) should be the same w

Re: [R] survreg with measurement uncertainties

Hi Terry, Thanks for your quick reply. I am talking about uncertainty in the response. I have 2 follow up questions: 1) my understanding from the documentation is that 'id' in cluster(id) should be the same when the predictors are not independent. Is this correct? (To be more concrete: my dat

Re: [R] survreg with measurement uncertainties

I will assume that you are talking about uncertainty in the response. Then one simple way to fit the model is to use case weights that are proprional to 1/variance, along with +cluster(id) in the model statement to get a correct variance for this case. In linear models this would be called the

Re: [R] survreg with measurement uncertainties

survreg allows interval censored data, if that is how you want to represent measurement uncertainty. See ?Surv -Original Message- From: Kyle Penner [mailto:kpen...@as.arizona.edu] Sent: Tuesday, June 11, 2013 8:02 PM To: r-help@r-project.org Subject: [R] survreg with measurement

[R] survreg with measurement uncertainties

Hello, I have some measurements that I am trying to fit a model to. I also have uncertainties for these measurements. Some of the measurements are not well detected, so I'd like to use a limit instead of the actual measurement. (I am always dealing with upper limits, i.e. left censored data.)