Example Matlab Simulation

1
Example Matlab Simulation
2
Based on Teaching Social Simulation with
MatlabJournal of Artificial Societies and Social
Simulation vol. 3, no. 1, ?lthttp//www.soc.surrey.
ac.uk/JASSS/3/1/forum/1.htm
3
"Take it or leave it."

• A friend is looking for a mate
• Goal Find matches that meet criterion number of
  desirable attributes in a mate
• honesty, intelligence, beauty, lets you cheat off
  them
• Different people have different numbers of these
  attributes - we give them a score based on the
  number of desirable attributes they have. - we
  can describe the population by a distribution on
  this score.
• Picky friend makes a wants a 12.
• How hard will that be to achieve?

4
Prospect distribution

• Prospects scores are normally distributed and
  have mean 6, standard deviation 3.
• Scores can be negative, but all scores are
  integers.

Meanscore 6 Stdscores 3 NumberProspects
1000 prospectsceil(63randn(1,NumberProspects))
This is your dating pool
5
How many prospects for friend?
goodonesfind(prospects gt 12)
6
Now what?

• Which dates should the friend take?
• How many possible matches?
• That depends on the PROSPECTs criteria
• We need to know our friends score. Say it is a
  10
• Assign each PROSPECT a criteria.
• Let us say PROSPECT criteria are about /-2
  points from their own score on average.
• Finally MATCHES are made first date that both
  exceed criteria.

7
PROSPECT CRITERIA
prospectcriteriaceil(prospects2randn(1,NumberPr
ospects))
8
MATCH GAME
Isdatablecount 0 For possibledates1NumberPro
spects, isdatable prospects(possibledates)gtfri
endscriterion isdatablecount isdatablecount
1 if isdatable, ismatch friendsscore gt
prospectscriterion(possibledates if
ismatch, matchscore prospects(possibledates)
break end end end
9
Be systematic!

• What is the average match scores
• As a function of friends score?
• What is the effect of
• Dating pool size
• Pool quality