[R] Maximum likelihood estimation in R with censored Data

[Vincent Goulet]

Le ven. 13 juin à 13:55, Ben Bolker a écrit :

> Bluder Olivia <olivia.bluder <at> k-ai.at> writes:
>
>>
>> Hello,
>>
>> I'm trying to calculate the Maximum likelihood estimators for a  
>> dataset
>> which contains censored data.
>>
>> I started by using the function "nlm", but isn't there a separate  
>> method
>> for doing this for e.g. the "weibull" and the "log-normal"  
>> distribution?
>>
>> Thanks,
>>
>> Olivia
>
>  This is not *quite* enough detail about what you
> want to do.  Can you (as the posting guide suggests!)
> give us a small example of what you want to do?  You may be able
> to do this via the survreg() command in the survival
> package, or you may want to do it yourself by constructing
> a log-likelihood function with dweibull() for uncensored
> data and pweibull() for censored data [or dlnorm/plnorm].

If you want to go the second route, function coverage() in package  
actuar will build the censored density function for you. You can then  
feed this function to fitdistr() just like for "usual" ML estimation.

HTH  Vincent

>
>
>  Ben Bolker
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.