Hi just wondering if there is a package that can get the maximum likelihood or method of moments estimator for distributions with censored data? The distributions I'm interested in are: Exponential, pareto, beta, gamma and lognormal. -- View this message in context: http://www.nabble.com/MLE-for-censored-distributions-in-R-tp15022863p15022863.html Sent from the R help mailing list archive at Nabble.com.
> Hi just wondering if there is a package that can get the maximum likelihood > or method of moments estimator for distributions with censored data? The > distributions I'm interested in are: Exponential, pareto, beta, gamma and > lognormal.Look at the survreg function in the survival library. It can find the MLE for any distribution for which g(y) can be written in a location-scale form, for some transform g. See names(survreg.distributions) for a list of those that are built in; others can be added. Terry Therneau
Le mar. 22 janv. ? 17:47, Thomas Downey a ?crit :> > Hi just wondering if there is a package that can get the maximum > likelihood > or method of moments estimator for distributions with censored > data? The > distributions I'm interested in are: Exponential, pareto, beta, > gamma and > lognormal.I'm highly biased here, but you may have a look at package actuar. The function coverage() will define the pdf or cdf of a random variable modified in many sorts of ways (truncation, censoring, inflation, etc.) based on any distribution you want. You can then use the function is usual fitting procedures such as fitdistr() for ML. See the "lossdist" vignette in the package for details. HTH --- Vincent Goulet, Associate Professor ?cole d'actuariat Universit? Laval, Qu?bec Vincent.Goulet at act.ulaval.ca http://vgoulet.act.ulaval.ca