# Translating subsets in normed spaces

Let $X$ be a Banach space and endow the space $BC(X)$, the space of all bounded closed subsets of $X$, with the Hausdorff distance $d_H$. Fix $C_0\in BC(X)$. Is it true that $d_H(A,B)=d_H(A+C_0,B+C_0)$ for all $A,B\in BC(X)$?

Edit: Is it true when $A,B,C_0$ are convex sets?

-

Counterexample in $\mathbb R$: $A=[-1,1]$, $B=\{-1,1\}$, $C_0=[-1,1]$. Here $d_H(A,B)=1$ but $A+C_0=B+C_0=[-2,2]$. Did you want the sets to be convex, by any chance?

Okay, let's consider the convex case. Let $X$ be a real Banach space; if it's complex, forget the complex structure. Let $S$ be the unit sphere of the dual space $X^*$. Given a bounded closed set $A\subset X$, define its support function $h_A\colon S\to\mathbb R$ by $h_A(\phi)=\sup_{A}\phi$. The definition implies $h_{A+C}=h_A+h_C$ for any bounded closed sets $A,C$. In particular, the quantity $\rho(A,B)=\sup_{S}|h_A-h_B|$ is invariant under addition/translation: $\rho(A+C,B+C)=\rho(A,B)$.

So far so good, but what does $\rho$ have to do with the Hausdorff metric? The inequality $\rho(A,B)\le d_H(A,B)$ holds for general $A,B\in BC(X)$: just use the definition of $d_H$ together with the fact that $\phi$ has norm $1$. The reverse inequality is where we need convexity... and it seems that we need $X$ to be a Hilbert space too... (?)

So, suppose $X$ is Hilbert and $A$ and $B$ are convex. We can find a point $p$ in one of these sets (say, in $A$) which is at distance $>d_H(A,B)-\epsilon$ from the other set (namely, $B$). There exists $q\in B$ that minimizes the distance to $p$ (a closed convex set in a Hilbert space has a unique nearest point to any point of its complement). Let $\phi$ be the norming functional of $p-q$, simply $\phi=(p-q)/\|p-q\|$. Then $h_A(\phi)-h_B(\phi)>d_H(A,B)-\epsilon$ as required.

-

$\newcommand{\ep}{\varepsilon}$$\newcommand{\f}{\phi} NOTE: this is an extension of the answer given by Leonid Kovalev. I continu with the notation from his answer. The Hausdorff distance is given by$$d_h(A,B)=\inf\{\ep>0\colon A\subset B+B_\ep(0)\text{ and }B\subset A+B_\ep(0)\},$$where B_\ep(0) is the open ball of radius \ep centered at 0 (see here). The identity d_h(A,B)=\sup_{\phi\in S}h_A(\phi)-h_B(\phi)| holds for any convex, bounded closed, nonvoid subset of a Banach space X (It need not be Hilbert) First note that if A=B, then the result is clear. Now assume A\neq B and let \ep>0 such that A\subset B+B_\ep(0). Then h_A(\f)\le h_B(\f)+\ep and h_B(\f)\le h_A(\f)+\ep for all \f\in S. Thus |h_A(\f)-h_B(\f)|<\ep, hence \sup_{\f\in S}|h_A(\f)-h_B(\f)|\le d_h(A,B). Next, define$$\ep:=\sup_{\f\in S}|h_A(\f)-h_B(\f)|>0,$$then$A\subset B+B_\ep(0)$and$B\subset A+B_\ep(0).$Thus$d_h(A,B)\le\ep\$.

-