[R] Permuting friendship nominations in a social network

[Jeff Newmiller]

It might be, but with appropriate indexes a SQL engine (via sqldf or RODBC for example) might be able to do it that way anyway.
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.

hollymaya <hollymaya at gmail.com> wrote:
>Jean, 
>Thank you for the suggestion. Actually the dataset is quite large so
>that method might be unmanageable. 
>Holly
>
>
>hollymaya at gmail.com
>
>
>
>On Sep 4, 2013, at 10:14 AM, "Adams, Jean" <jvadams at usgs.gov> wrote:
>
>> Holly,
>> 
>> I don't know of a clever way to do this, but I can think of a brute
>force way, which might only be feasible if you have a small data set
>(as in your example).  You could permute every possible set of
>connections, then choose from that collection only the ones that meet
>your criteria.  
>> 
>> Using your example, there are c=21 possible connections among the n=7
>unique individuals, c = n*(n-1)/2.  Your example shows a total of 8
>connections (16 rows / 2).  So you could generate all permutations of
>choose(21, 8) = 203,490 ways to have 8 connections.  Then subset the
>ones that have individual totals the same as your example (1 connection
>for Alicia and Beth, 2 for Kerry and Kim, 3 for James and John, and 4
>for Rachel).
>> 
>> Jean
>> 
>> 
>> On Tue, Sep 3, 2013 at 1:01 PM, hollymaya <hollymaya at gmail.com>
>wrote:
>> 
>> I have a dataset of dyads (an edgelist) representing friendship
>nominations between egos and their nominated alters. The network is
>undirected so if ego is connected to alter, then there is a separate
>observation in the dataset for the reverse. I would like to randomly
>permute the friendships so that 1.) the total degree for each
>individual remains the same, i.e. each individual ends up with the same
>number of friendships they had in the original undirected dataset and
>2.) there are no self loops, so individuals are not connected to
>themselves. Any suggestions on this would be greatly appreciated.
>> 
>> 
>> 
>> Example
>> 
>> Observed data:
>> 
>> Ego          Alter
>> 
>> Alicia       James
>> 
>> Beth        Kim
>> 
>> James      John
>> 
>> James      Rachel
>> 
>> James      Alicia
>> 
>> John        Kerry
>> 
>> John        Rachel
>> 
>> John        James
>> 
>> Kerry     Rachel
>> 
>> Kerry      John
>> 
>> Kim         Rachel
>> 
>> Kim          Beth
>> 
>> Rachel    Kim
>> 
>> Rachel    James
>> 
>> Rachel    Kerry
>> 
>> Rachel    John
>> 
>> 
>> 
>> Permuted data:
>> 
>> Ego          Alter
>> 
>> Alicia
>> 
>> Rachel
>> 
>> Beth
>> 
>> James
>> 
>> James
>> 
>> Beth
>> 
>> James
>> 
>> John
>> 
>> James
>> 
>> Kim
>> 
>> John
>> 
>> Rachel
>> 
>> John
>> 
>> Kerry
>> 
>> John
>> 
>> James
>> 
>> Kerry
>> 
>> Rachel
>> 
>> Kerry
>> 
>> John
>> 
>> Kim
>> 
>>  Rachel
>> 
>> Kim
>> 
>> James
>> 
>> Rachel
>> 
>> Kim
>> 
>> Rachel
>> 
>> John
>> 
>> Rachel
>> 
>> Alicia
>> 
>> Rachel
>> 
>> Kerry
>> 
>> 
>> 
>> Thank you in advance,
>> Holly
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> hollymaya at gmail.com
>> 
>> 
>> 
>> 
>>         [[alternative HTML version deleted]]
>> 
>> ______________________________________________
>> 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.
>> 
>
>
>	[[alternative HTML version deleted]]
>
>______________________________________________
>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.