[R] Fw: Estimating Parameters of Weibull and Pareto distribution using LMOM package
J. R. M. Hosking
JRMH001 at gmail.com
Wed Mar 18 21:25:41 CET 2009
Maithili Shiva wrote:
> Dear R helpers
>
> I have r file which estimate the parameters of 3 parameter
> Weibull -
>
> (A) - continuous shape parameter (alpha)
> - continuous scale parameter (beta)
> - continuous location parameter (gamma)
>
> (B) Also, I have a r file which calculates the parameters
> of Generalized Pareto distribution.
>
> - location parameter xi,
> - scale parameter alpha and
> - shape parameter k
>
>
> However, If I have to use the same files for estimating
> parameters of 2 parameter Weibull and 2 parameter Pareto distribution, how do I use it?
>
>
> I am giving the R script I am using to calculate the Generalized Pareto distribution as
>
> library(lmom)
>
> amounts <- (10023.47, 10171.42,13446.83,10263.49,10219.07, 10025.71, 10318.88, 10034.85,10004.98,10012.72)
>
> lmom <- samlmu(amounts); lmom
>
> parameters_of_Gen_Pareto <- pelgpa(lmom);
> parameters_of_Gen_Pareto
>
>
> The parameters estimated are
>
> xi alpha k
> 9993.3131812 81.9540457 -0.8213843
>
>
> If the location paramter xi = 0, then this becomes two
> parameter Paretom distribution. However, it is my gut
> feeling that if xi = 0, other parameter values will also
> change.
pelgpa allows you to specify the lower bound of the distribution and
estimate the other two parameters. Compare
> pelgpa(samlmu(amounts))
xi alpha k
9993.3131812 81.9540457 -0.8213843
and
> pelgpa(samlmu(amounts),bound=10000)
xi alpha k
10000.000000 72.993343 -0.838561
pelwei offers similar options for the Weibull distribution.
J. R. M. Hosking
> Please help me.
>
> Regards and thanking in advance
>
> Maithili
>
> ______________________________________________
> 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.
>
More information about the R-help
mailing list