st: Challenging problem.

["Victor Zammit" <vmz@go.net.mt>]

Dear Stata Users,

The following is a table,representing the results five counts in a local 
election with 1352 valid ballots.
The electing quota for a candidate was 320.Candidate #3,can_3,is elected in 
the first count,c1,
with first preference.The extra votes,721-320 = 401,get distributed 
proportionately to
the candidates that receive second preference,after can_1's (first 
preference) votes.
can c1 c2 c3 c4 c5

can_1 295 423 423 320 320

can_2 20 39 58 90 0

can_3 721 320 320 320 320

can_4 253 442 320 320 320

can_5 63 102 203 270 356

n-t.v. 0 26 28 32 36

total 1352 1352 1352 1352 1352

Subtracting c1 from c2, gives the proportinate amounts that were distributed 
to the other
candidates and the proportionate amount of the non-transferable 
votes,n-t.v.,from which,can be
derived the second prefernce that each candidate and non transferables 
,receive.
pref_2 = [[ c2-c1]/401] * 721, format pref_2 %3.0f .

can c1 c2 p_dist pref_2

can_1 295 423 128 230

can_2 20 39 19 34

can_3 721 320 0 0

can_4 253 442 189 340

can_5 63 102 39 70

n-t.v. 0 26 26 47

The extra over 320 for can_4 and after that,can_1,are dealt with 
similarly.In the fourth count,
the accumulation of 90 votes for can_2,are distributed directly on can_5 and 
n-t.v. .
Given this information,is it possible to simulate the 1352 ballots?.I 
appreciate any feedback..
Victor Zammit

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/