Let $f:\mathbb{R}^2 \to \mathbb{R}$ such that $f(x_1,x_2)=x_2$. I am required to show that $$f(U)= \big\{ f(x_1,x_2) : (x_1,x_2) \in U \big\}$$ is an open set of $\mathbb{R}$ with the standard metric. Given that U is an open subset of $\mathbb{R}^2$. The metric on $\mathbb{R}^2$ is given by $$d((x_1,x_2),(y_1,y_2)):= \max \big\{\vert x_1-y_1\vert,\vert x_2-y_2\vert \big\}.$$ Can anyone please give me any hints with how to proceed with this question?

  • $\begingroup$ How is $U$ given? $\endgroup$ – Hetebrij Apr 7 '16 at 17:29
  • $\begingroup$ Is $U$ an open set in$\mathbb{R}^2$. ? $\endgroup$ – seeker Apr 7 '16 at 17:29
  • $\begingroup$ The metric on $\mathbb{R}^2$ is not well-defined. $\endgroup$ – sqtrat Apr 7 '16 at 17:30
  • $\begingroup$ Note that $f((a,b) \times (c,d)) = (c,d)$. Every point $x$ in $U$ is contained in some rectangle $(a,b) \times (c,d) \subset U$. $\endgroup$ – copper.hat Apr 7 '16 at 17:33
  • $\begingroup$ @seeker Yes , sorry $\endgroup$ – anon Apr 7 '16 at 17:42

Suppose that $U$ is open in $\mathbb{R}^2$. If $f(U) = \emptyset$, we are done. So assume that $f(U)\not=\emptyset$. If $x_2 \in f(U)$, then there is an $(x_1,x_2)$ in $U$ for some $x_2$. Since $U$ is open there is an $\varepsilon >0$ such that $V:= (x_1-\varepsilon, x_1+\varepsilon)\times (x_2-\varepsilon, x_2+\varepsilon) =B_d((x_1,x_2),\varepsilon)\subset U$. Then $$(x_2-\varepsilon,x_2+\varepsilon) = f(V)\subset U$$ which is clearly an open subset of $\mathbb{R}$.

  • $\begingroup$ Thanks for your reply. Can we just state that $f(V)=(x_2 - \epsilon, x_2 + \epsilon)$ ? $\endgroup$ – anon Apr 7 '16 at 17:51
  • $\begingroup$ Yes, because $f$ is the projection map onto the second factor and the balls in the space under the metric $d$ are open squares. $\endgroup$ – sqtrat Apr 7 '16 at 17:54

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.