2
$\begingroup$

I'd like to verify the following equality in Mathematica:

$$ P(Z_1+Z_2=2,Z_1=1,Z_2=1)=P(Z_1=1,Z_2=1) $$

by knowing that $Z_1,Z_2 \in \lbrace 1,2,3,4 \rbrace$ are two independent and uniform random variables.

$\endgroup$
9
  • $\begingroup$ What should P(z1+z2==2,z1==1,z2==1) mean? $\endgroup$ Jan 8, 2022 at 14:17
  • $\begingroup$ Hello @DanielHuber the commas mean intersections $\endgroup$ Jan 8, 2022 at 14:37
  • 1
    $\begingroup$ But in this case P(z1+z2==2,z1==1,z2==1) is a tautology, z1==1&& z2==1 imply z1+z2==2 $\endgroup$ Jan 8, 2022 at 14:43
  • $\begingroup$ @DanielHuber yes, is it possible to simplify the left term in Mathematica? The simplified one should be the right term $\endgroup$ Jan 8, 2022 at 14:45
  • $\begingroup$ You may e.g. write your equations by zeros of polynomials and use Groebner Basis like: GroebnerBasis[{z1 + z2 - 2, z1 - 1, z2 - 1}, {z1, z2}] Or you introduce a third variable and eliminate it like: Eliminate[{z == z1 + z2 - 2, z1 == 1, z2 == 1}, z] $\endgroup$ Jan 8, 2022 at 14:55

2 Answers 2

5
$\begingroup$

The code below calculates the two probabilities and allows you to verify that they are the same by inspection.

Probability[
 {z1 == 1 && z2 == 1, z1 == 1 && z2 == 1 && z1 + z2 == 2},
 {
  Distributed[z1, DiscreteUniformDistribution[{1, 4}]],
  Distributed[z2, DiscreteUniformDistribution[{1, 4}]]
 }
]

(*Out: {1/16, 1/16} *)
$\endgroup$
5
  • $\begingroup$ Hello @MarcoB thank you very much for your helpful code. $\endgroup$ Jan 8, 2022 at 18:41
  • $\begingroup$ How did you specify the independence of the random variables? $\endgroup$ Jan 8, 2022 at 20:05
  • $\begingroup$ @Gennaro They should be independent by default here. I think you'd have to work harder to indicate if they are somehow correlated. $\endgroup$
    – MarcoB
    Jan 8, 2022 at 20:09
  • $\begingroup$ if possible can you post the documentation where I can find this implicit hypothesis please? $\endgroup$ Jan 8, 2022 at 20:39
  • $\begingroup$ I found it here reference.wolfram.com/language/ref/Probability.html $\endgroup$ Jan 8, 2022 at 20:57
5
$\begingroup$

The answer by @MarcoB is almost always the way to go. But if the set of equations ( {z1 == 1 && z2 == 1, z1 == 1 && z2 == 1 && z1 + z2 == 2}) is complicated enough such that Probability doesn't work, you might try a brute force method.

This approach simply creates the sample space of equally likely outcomes and then selects the outcomes that satisfies each condition.

sampleSpace = Flatten[Table[{z1, z2, z1 + z1}, {z1, 1, 4}, {z2, 1, 4}], 1]
(* {{1, 1, 2}, {1, 2, 2}, {1, 3, 2}, {1, 4, 2}, {2, 1, 4}, {2, 2, 4}, 
    {2, 3, 4}, {2, 4, 4}, {3, 1, 6}, {3, 2, 6}, {3, 3, 6}, {3, 4, 6},
    {4, 1, 8}, {4, 2, 8}, {4, 3, 8}, {4, 4, 8}} *)

subset1 = Select[sampleSpace, #[[1]] + #[[2]] == 2 && #[[1]] == 1 && #[[2]] == 1 &]
(* {{1, 1, 2}} *)

subset2 = Select[sampleSpace, #[[1]] == 1 && #[[2]] == 1 &]
(* {{1, 1, 2}} *)

So the probabilities are both 1/16:

Length[subset1]/Length[sampleSpace]
(* 1/16 *)
Length[subset2]/Length[sampleSpace]
(* 1/16 *)
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.