0

Let $(X, d_x)$ and $(Y, d_y)$ be metric spaces. I was given distance functions:

$d_1((x_1,y_1),(x_2,y_2)) = d_x(x_1, x_2)+d_y(y_1,y_2)$ $d_\infty ((x_1,y_2), (x_2, y_2) = max \{d_x(x_1, x_2), d_y((y_1, y_2)\}$.

I'm trying to prove that these are metric spaces by proving they meet the three axioms (positivity, symmetry and triangle law). The first two conditions are quite trivial. I'm having difficulty proving the third one, where $d(x,y) \leq d(z,y) + d(x,z)$. Help would be very much appreciated.

Nikitau
  • 1,353

1 Answers1

1

It's just a matter of writing out all the notation, starting from the complicated part and manipulating it into the simpler part. The proof itself is pretty straightforward. Given any $x_1, x_2, x_3 \in X$ and any $y_1, y_2, y_3 \in Y$, notice that: \begin{align*} ~~&d_1((x_1, y_1), (x_3, y_3)) + d_1((x_3, y_3), (x_2, y_2)) \\ &= [d_x(x_1, x_3) + d_y(y_1, y_3)] + [d_x(x_3, x_2) + d_y(y_3, y_2)] \\ &= [d_x(x_1, x_3) + d_x(x_3, x_2)] + [d_y(y_1, y_3) + d_y(y_3, y_2)] \\ &\geq d_x(x_1, x_2) + d_y(y_1, y_2) \\ &= d_1((x_1, y_1), (x_2, y_2)) \end{align*} as desired. $~~\blacksquare$

Adriano
  • 41,576
  • Ah, thank you so much! For some reason I was "stuck" trying to justify line 3 and 4 ... and then I realized I was being very silly. Thank you! – Nikitau Oct 06 '16 at 03:54