From Wikipedia:

Let $F, G : \mathbb{R} \to [0, 1]$ be two cumulative distribution functions. Define the Lévy distance between them to be :$$L(F, G) := \inf \{ \varepsilon > 0 | F(x - \varepsilon) - \varepsilon \leq G(x) \leq F(x + \varepsilon) + \varepsilon \mathrm{\,for\,all\,} x \in \mathbb{R} \}.$$

I was wondering what is the role of $\varepsilon$ in the range values ? It appears weird to me to take the same $\varepsilon$ in the range and domain values.

Does the following definition still satisfies metric properties ?

$$L(F, G) := \inf \{ \varepsilon > 0 | F(x - \varepsilon) \leq G(x) \leq F(x + \varepsilon) \mathrm{\,for\,all\,} x \in \mathbb{R} \}.$$



$$K(F,G) := \inf\{\varepsilon>0; G(x-\varepsilon) \leq F(x) \leq G(x+\varepsilon) \, \text{for all} \, x \in \mathbb{R}\}.$$

It is not difficult to see that $K$ is a metric. Let us consider the following easy example in order to see the differences between the Lévy metric $L$ and the metric $K$:

Define random variables $X:= 1_{[1/2,1]}$ and $X_n := \frac{1}{2} 1_{(0,1/n]}+1_{[1/2,1]}$ on the unit interval $[0,1]$ (endowed with the Lebesgue measure). The corresponding distribution functions are given by

$$G(x) := \mathbb{P}(X \leq x) = 1_{[1,\infty)}(x) \qquad \qquad G_n(x) := \mathbb{P}(X_n \leq x) = \frac{1}{n} 1_{[1/2,1)}(x) + 1_{[1,\infty)}(x).$$

Using the definition of $L$ and $K$ it is not difficult to see that $L(G_n,G) \to 0$ whereas $K(G_n,G)=1/2$. On the other hand, we know that $X_n \to X$ almost surely; that's why the Lévy metric $L$ is more natural: If the random variables converge almost surely, then we would expect that the distance of the corresponding distribution functions converges to $0$.

A typical $\varepsilon$-neighborhood of the distribution function $G$ with respect to the Lévy metric $L$ looks like that

enter image description here

whereas a neighborhood with respect to $K$ is of the following form

enter image description here

Remark In fact, one can show that $L(G_n,G) \to 0$ if and only if $X_n \to X$ in distribution for any random variables $(X_n)_n$ and $X$.

  • $\begingroup$ Thanks a lot for your detailed answer. I understand that it is a matter of convergence. However I'm still interested in proving that the mapping K is a metric. Hereafter an attempt of the triangle inequality (the others are quiet obvious). We have to prove the following. $$ L(F,H)\leq L(F,G)+L(G,H) $$Consider $L(F,G)\leq a$ and $L(G,H)\leq b$ We have: \begin{eqnarray} G(x-a)\leq &F(x)& \leq G(x+a), \forall x\\ H(x-b)\leq&G(x)&\leq H(x+b) \forall x\\ H(x-a-b) \leq &F(x)& \leq H(x+a+b), \forall x \end{eqnarray} Then $L(F,H)\leq a+b$ $\endgroup$
    – user140328
    Apr 11 '14 at 8:04
  • $\begingroup$ @user140328 You have to assume that $L(F,G) < a$ and $L(G,H)<b$ (because, by definition, the inequality $$G(x-a) \leq F(x) \leq G(x+a)$$ holds for any $a>L(F,G)$ but not necessarily for $a=L(F,G)$). Then letting $a \downarrow L(F,G)$ and $b \downarrow L(G,H)$ finishes the proof. $\endgroup$
    – saz
    Apr 11 '14 at 8:30
  • $\begingroup$ Right ! Thanks. $\endgroup$
    – user140328
    Apr 11 '14 at 9:25

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.