I'm not sure of the correct notation if someone could please help.

Say you wish to have a set comprised of the union of two sets, such as $$h=\Big\{ \begin{bmatrix}x & y\\y & x\end{bmatrix}: x^2-y^2=1 \Big\};$$$$g=\Big\{ \begin{bmatrix}x & y\\-y & -x\end{bmatrix}: x^2-y^2=1 \Big\}.$$

I'm not sure if putting the union sign inside the set is correct. Would it be $$\phi = \Big\{ \begin{bmatrix}x & y\\y & x\end{bmatrix}\cup \begin{bmatrix}x & y\\-y & -x\end{bmatrix}\ : x^2-y^2=1 \Big\}?$$

  • $\begingroup$ $\phi=h\cup g$ or replace the $\cup$ in your final line by a comma. $\endgroup$ – Jonathan Feb 15 '14 at 17:59
  • 1
    $\begingroup$ Thanks, that's all I needed to know :) $\endgroup$ – user112633 Feb 15 '14 at 18:00
  • $\begingroup$ Ignore the example, it's just made up. $\endgroup$ – user112633 Feb 15 '14 at 18:01

You can

$(1)$: Provided you've already defined $h, g$, as you did in this post, the simplest route is to express $\phi$ as the union of $h, g$, as in $$\phi = h\cup g$$

$(2)$ Write $$\phi = \left\{ \begin{bmatrix}x & y\\y & x\end{bmatrix}\text{ or }\; \begin{bmatrix}x & y\\-y & -x\end{bmatrix}\ : x^2-y^2=1 \right\}$$

The union operation is an operation on sets; for example, $h, g$ are sets.

Instead of "or", you can use the notation $\lor$ in $(2)$.

Better yet, if you really like set-builder notation, you can write:

$$\phi = \{A: A \in g \lor A \in h\}$$ which means: "The set of all matrices A that belong to $h$ or belong to $g$.

You might also want to include the field to which $x, y$ belong. E.g., $x, y \in \mathbb R$.

  • $\begingroup$ So you wouldn't include set brackets in the case of (1)? $\endgroup$ – user112633 Feb 15 '14 at 18:07
  • $\begingroup$ That's exactly right. Once $g, h$ are defined, we can express $\phi$ as the union of two sets, which is, again, a set. $\endgroup$ – amWhy Feb 15 '14 at 18:09
  • $\begingroup$ @amWhy: Have you seen that notation in $(2)$ used before? I think I would find it confusing if it weren't in context that made it obvious what was meant. $\endgroup$ – user14972 Feb 15 '14 at 18:10
  • $\begingroup$ @amWhy Thank you, I would upvote, but it won't let me log in to my account; probably because I am at uni. I will do later. $\endgroup$ – user112633 Feb 15 '14 at 18:12
  • $\begingroup$ I would not use $(2)$. If using set-builder notation, I'd use $$\phi = \{A: A \in g \lor A \in h\}$$ which essentially means $\phi = g\cup h$. $\endgroup$ – amWhy Feb 15 '14 at 18:12

Maybe $$ \left\{\,\begin{bmatrix}x&y\\ex&ey\end{bmatrix}\in \mathbb R^{2\times 2}: x^2-y^2=1, e=\pm1\,\right\}$$ This matches the set-builder syntax $\{\,x\in X: \phi(x)\,\}$ as per Axiom schema of Specification (or Comprehension). Depending on context, the class-builder syntax $\{\,x:\phi(x)\,\}$ may be ok, but at any rate having a "complex expression" to the left of the colon is at least ambiguous as it may suggest a replacement $\{\,f(x):x\in X\,\}$. (Actually, the matrix itseld is already such a "complex expression", so this is arguably an instance of replacement anyway).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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