Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm trying to formulate Axiom of Choice in terms of $P(\prod_{i \in I}{X_i})$, but end up with the following questions.

Let $P(X)$ be the power set of $X$.In general, is it correct, or legitimate to write: $P(\prod_{i \in I}{X_i}) = \prod_{i \in I}{P(X_i)}$?

Let $|X|$ be the cardinality of $X$.Is it the case $|P(\prod_{i \in I}{X_i})| \leq \prod_{i \in I}{|P(X_i)|}$?

share|cite|improve this question
up vote 4 down vote accepted

Note that all the elements in $\mathcal P(\prod X_i)$ are sets of functions, whereas $\prod\mathcal P(X_i)$ is a set of functions whose range is sets.

So equality never really holds.

In terms of cardinality, Ross already showed the inequality fails for finite sets. However assuming the axiom of choice we can say the following: Denote by $\mathcal P^+(X)$ all the non-empty subsets of $X$, then: $$\left|\prod_{i\in I}\mathcal P^+(X_i)\right|\leq\left|\mathcal P\left(\prod_{i\in I} X_i\right)\right|$$

This is given by the function $\langle A_i\mid i\in I\rangle\mapsto\{g\mid \forall i\in I. g(i)\in A_i\}$ is clearly injective.

The reason we need to consider $\mathcal P^+(X_i)$ is that if the empty set is present in more than one factor the above function is not an injection. Depending on the cardinality of $I$ and the $X_i$'s it might be possible to show there exists an actual injection from the product of the power sets.

Without the axiom of choice, however, things may be different. If $X_i\neq\varnothing$ but $\prod X_i=\varnothing$ then we have $\mathcal P(\prod X_i)=\{\varnothing\}$ whereas $\prod\mathcal P(X_i)$ is never empty because the function $f(i)=X_i$ is in that product (as well in the product $\prod\mathcal P^+(X_i)$).

It is even easy to show that if at least one $X_i$ has two elements then $\prod\mathcal P^+(X_i)$ has more than one element (let alone the product of the full power sets).

It might be nicer, if so, to formulate the axiom of choice as the inequality above about the product of $\mathcal P^+(X_i)$ having an injection into the power set of the product. Although to be fair I'm not sure I see much point to such formulation.

share|cite|improve this answer
Thank you for your answer. I'm more curious about the situation that $X_i$ is non-empty, while $\prod_{i \in I}{X_i}$ is empty. Is it why you use the phrase "almost true"? – Metta World Peace Jan 30 '13 at 11:28
Metta, I began writing a comment but it got long and detailed. I will just edit it into my answer. – Asaf Karagila Jan 30 '13 at 13:17

They are not equal. If $X_1=\{a,b,c\}, X_2=\{d,e,f\}$ the left side is the power set of a nine-element set whose elements are ordered pairs and has size $2^9=512$. The right side is a set of ordered pairs of the subsets of two three-element sets and has size $2^3 \cdot 2^3=2^6=64$.

share|cite|improve this answer
Drats, you got ahead of me by a mere minute! Oh well. :-) – Asaf Karagila Jan 29 '13 at 16:09
Ross, while the first comment showed that the statement itself is not true and therefore cannot be equivalent to the axiom of choice; the question seems to me as asking "Is this statement equivalent to the axiom of choice"? Your edit, if so, is irrelevant to that question. – Asaf Karagila Jan 29 '13 at 16:16

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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