[R] Ranked Set Sampling

[David Winsemius]

On Oct 3, 2010, at 3:01 PM, David Winsemius wrote:

>
> On Oct 3, 2010, at 2:13 PM, Jeffrey Spies wrote:
>
>> This is certainly not my area of expertise, but like Peter mentioned,
>> Jeff Terpstra published this:
>>
>> http://www.jstatsoft.org/v14/i07
>>
>> which has R code listed as supplements.  Joe McKean seems to keep an
>> updated version of that code here:
>>
>> http://www.stat.wmich.edu/red5328/WWest/
>>
>> And Brent Johnson has extended that code for variable selection/ 
>> regression:
>>
>> http://userwww.service.emory.edu/~bajohn3/software.html
>>
>> Maybe that's a start; if you find more by following Peter's  
>> suggestion
>> of privately contacting authors, please follow-up with the list.
>
> Yes; search terms ' "ranked set sampling" "r-project" ' ... does  
> produce a very complete book-length piece: "Robust Nonparametric  
> Statistical Methods" (2010) by Hettmansperger and McKean which has a  
> section on the question posed at printed pg 53 (pg 64 of the pdf  
> version) and on second print page cites R code available:
>
> http://www.stat.wmich.edu/mckean/Rfuncs/

And that link does not succeed, but  this one does:

http://fisher.stat.wmich.edu/joe/Stat666/Rfuncs/

>
> (And I note that Terpstra and McKean are co-authors and that the  
> same institution.)
>
>>
>> Cheers,
>>
>> Jeff.
>>
>> On Sun, Oct 3, 2010 at 1:38 PM, Peter Dalgaard <pdalgd at gmail.com>  
>> wrote:
>>> On 10/03/2010 06:32 PM, David Winsemius wrote:
>>>>
>>>>
>>>> Ahmed Albatineh wrote:
>>>>>
>>>>>
>>>>> Are you aware of any package that calculates Ranked Set Sample?  
>>>>> If you
>>>>> have
>>>>> a code that you are willing to share, I will acknowledge that in  
>>>>> my work.
>>>>> Thanks much
>>>>>
>>>>> Ahmed
>>>>>
>>>>
>>>> I wonder if this is a phrase that is uniformly understood? One  
>>>> possibility
>>>> is that you are asking to sample elements of a set based on some  
>>>> ranking
>>>> function. In that case you may need to describe in more detail  
>>>> how you want
>>>> to handle ties and whether this function is supposed to deal with
>>>> multivariate strata. (There are many base functions that handle  
>>>> univariate
>>>> situations and there are packages that provide support for more  
>>>> complex
>>>> ones.)
>>>>
>>>> You are also requested (in the Posting Guide) to provide an  
>>>> example  that
>>>> can be cut and pasted and desired results against which responder  
>>>> can judge
>>>> the degree to which their efforts agree with your hopes.
>>>>
>>>
>>> It's a fairly well-defined concept. I.e., you can google for it...
>>>

David Winsemius, MD
West Hartford, CT