[R] lmomco package - Random number generation using Wakeby distribution

[David Winsemius]

On Jan 21, 2013, at 6:39 PM, David Winsemius wrote:

> 
> On Jan 21, 2013, at 6:15 PM, Katherine Gobin wrote:
> 
>> Dear Sir,
>> 
>> Thanks a lot for your eye-opener reply. I was just thinking of our usual commands like rnorm, runif etc. So I was wondering if there exists something like rwakeby etc. 
>> 
>> And lastly, I have calculated the parameters using 
>> 
>>> lmr                    = lmom.ub(amounts)
>>> parameters_of_Wakeby   = parwak(lmr)
>> 
>> whereas you have mentioned  lmom2par(), Will it create different set of parameters? Actually I am travelling and don't have R installed on the laptop I am carrying with me to verify ther results.
> 
> Neither this posting nor the first one had any data. I'm basically quoting the help files and making what I thought were sensible suggestions that were untested in the absence of data (and in this case in the absence of even code). I have no experience working with this package or with the Wakeby distribution.

I tested with the example offered on ?parwak : 


> lmr <- lmomco::lmom.ub(rnorm(20))
>  par.wakeby <-  lmomco::parwak(lmr)
> lmomco::rlmomco(10, par.wakeby)
 [1]  2.74458443  0.12585363  0.09981644 -0.72773835  0.67986712  0.02803862
 [7]  0.16152205 -0.62631478 -0.56486845  0.34771307


> 
> -- 
> David.
> 
>> 
>> Regards
>> 
>> Katherine
>> 
>> 
>> 
>> --- On Mon, 21/1/13, David Winsemius <dwinsemius at comcast.net> wrote:
>> 
>> From: David Winsemius <dwinsemius at comcast.net>
>> Subject: Re: [R] lmomco package - Random number generation using Wakeby distribution
>> To: "Katherine Gobin" <katherine_gobin at yahoo.com>
>> Cc: r-help at r-project.org
>> Date: Monday, 21 January, 2013, 7:46 PM
>> 
>> 
>> On Jan 21, 2013, at 10:30 AM, Katherine Gobin wrote:
>> 
>>> Dear R forum
>>> 
>>>> From the given data, I have estimated the parameters of Wakeby distribution using lmomco package as
>>> 
>>> library(lmomco)
>>> 
>>> (amounts <- read.csv("input_S.csv")$amount)
>>> 
>>> # ___________________________________________________________
>>> 
>>> # Wakeby distribution - Parameter estimation
>>> 
>>> N                      =
>>> length(amounts)
>>> lmr                    = lmom.ub(amounts)
>>> parameters_of_Wakeby   = parwak(lmr)
>> 
>> It appears you have a) not included the code that produced that output and b) failed to read the Index page for that package
>> 
>> help(package="lmomco")
>> 
>> help(package="lmomco")
>> 
>> ?rlmomco    #  Random Deviates of a Distribution
>> 
>> So on the assumption that you have an object in your workspace named "parameters_of_Wakeby" and it is an lmomco produced object like that returned by lmom2par() I would try:
>> 
>> rlmomco(100, parameters_of_Wakeby) 
>> 
>> 
>>> 
>>>> parameters_of_Wakeby
>>> 
>>> $type
>>> [1]
>>> "wak"
>>> 
>>> $para
>>>                  xi                alpha 
>>> 1.18813927666405e+04 0.00000000000000e+00 
>>>                beta                gamma 
>>> 0.00000000000000e+00 8.11391042554567e+04 
>>>               delta 
>>> 9.57554297149062e-01 
>>> 
>>> This means the scale parameters are 0.
>>> 
>>> However, assuming, all the five parameters of Wakeby distribution (viz. location parameter m (xi), the scale parameters a, b, and shape parameters  g and d are available. 
>>> 
>>> Then, how do I generate say 100 random no.s using Wakeby distribution w.r.t. these
>>> 5 available parameters.
>>> 
>>> I couldn't find any information about this in lmomco. Kindly guide if random no.s can be generated or not and if yes, how it can be done in r.
>> 
>> You should have been able to find this with:
>> 
>> help.search("random", package="lmomco")
>> 
>> -- 
>> 
>> David Winsemius
>> Alameda, CA, USA
>> 
> 
> 
> David Winsemius
> Alameda, CA, USA
> 
> ______________________________________________
> 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.

David Winsemius
Alameda, CA, USA