# [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

>
>  Regards and thanking in advance
>
>  Maithili
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help