MathGroup Archive 2005

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: can Mathematica be useful for this?

  • To: mathgroup at
  • Subject: [mg56576] Re: can Mathematica be useful for this?
  • From: "Kezhao Zhang" <kezhao.zhang at>
  • Date: Fri, 29 Apr 2005 03:20:50 -0400 (EDT)
  • References: <d4kks5$eck$>
  • Sender: owner-wri-mathgroup at

Here is another approach:

For each language, let's use 1 to indicate that the person can speaks
it and 0 that he cannot. So a binary vector of length 3 describes what
languages the person speaks. For example, {S=1, F=0, G=0}, or {1, 0, 0}
means that he can speak Spanish only. The probably corresponding to
such vector is denoted by p[S,F,G] (lower case p).  All the
probabilities listed in the problem are functions of p[S,F,G] (all 8 of
them). For example, probability that the person can speak S and F, P[S
and F]=p[1,1,0]+p[1,1,1] (note the difference between upper case P and
lower case p), etc.

We need to write P in terms of p and solve for p which is the answer we

The probabilities listed in the problem are



Now write each of them in terms of p:





The equation to be solved:


Plus the condition that all p's sum to 1:

Solve for p:

sol=Solve[Append[eqn, Total[unknown]==1],unknown]


Hence the probability that the person doesn't speak any is
p[0,0,0]=73/150, and the probability that he speaks one only is



K. Zhang

  • Prev by Date: Re: Re: NIntegrate and NDSolve
  • Next by Date: Re: Floor[] Not An Integer?
  • Previous by thread: Re: can Mathematica be useful for this?
  • Next by thread: MapThread Replace?