Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

A question in metric space Rn

0 views
Skip to first unread message

miki

unread,
Jul 4, 2009, 7:40:05 AM7/4/09
to
Hello all,

I'd be happy if someone here will be able to solve my problem.

Assume all metric spaces are Rn.

Let Y be a closed set of a metric space X.
Let x and z be some points in X (dont belong to Y)
then if

3 * r(x, z) < r(z, Y)

then

2 * r(x, z) < r(x, Y)

where r is the distance function. Moreover, I assume that the distance
between a point and a set is the infimum of all distances between the
point and set over the points of the set.

Any help would be highly appreciated.

Miki

Chip Eastham

unread,
Jul 4, 2009, 8:02:25 AM7/4/09
to

Hint: Argue by contradiction using the
triangle inequality. For example, if
there exists y in Y s.t.

r(x,y) <= 2*r(x,z)

what can you say about r(z,y) ? This
just uses properties of metric spaces.

regards, chip

TCL

unread,
Jul 4, 2009, 9:48:20 AM7/4/09
to

Use the inequality:
|r(z,Y)-r(x,Y)| <= r(x,z)

-TCL

miki

unread,
Jul 4, 2009, 11:02:18 AM7/4/09
to
> -TCL- Hide quoted text -
>
> - Show quoted text -

Well, using the inequality you have mentioned indeed solves the
problem. But, why is that inequality true?
How can I prove it?

Thanks a lot,
Miki


José Carlos Santos

unread,
Jul 4, 2009, 12:06:25 PM7/4/09
to

In order to prove that r(z,Y) - r(x,Y) <= r(x,z), I shall in fact prove
that r(z,Y) <= r(x,z) + r(x,Y). Let _y_ be an element of Y. Then

r(z,y) <= r(z,x) + r(x,y).

Since this is true for every _y_, it follows that

r(z,Y) <= r(x,z) + r(x,Y).

By symmetry, r(x,Y) - r(z,Y) <= r(x,z). Therefore,

|r(z,Y) - r(x,Y)| <= d(x,z).

Best regards,

Jose Carlos Santos

MeAmI.org

unread,
Jul 4, 2009, 5:50:29 PM7/4/09
to
MeAmI.org scribed:

Is it possible to adapt the a and b variables?

R and N?


Yes.
b = 95.951(6) , V = 1403.4(2) 3 , space g. P2/n, Z = 2,
crystal size 0.22 .... Metric parameters of (PNP)AlCl 2 are reproduced
to within 0.05 /3 ...h

0 new messages