Scribd
Upload
68 views2 pages

PMA307: Metric Spaces Solutions 5

This document contains solutions to problems about metric spaces. Problem 5.1 defines the inverse image of an annulus under a function f as the set of all points whose image under f lies within the annulus. Problem 5.2 shows that a function f from R2 to R is continuous, and defines the inverse image of an interval as the set of points mapping into that interval. Problem 5.3 provides a counterexample to show a function from R to R is not continuous by examining its inverse image of a closed set.

Uploaded by

Tom Davis
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
68 views2 pages

PMA307: Metric Spaces Solutions 5

This document contains solutions to problems about metric spaces. Problem 5.1 defines the inverse image of an annulus under a function f as the set of all points whose image under f lies within the annulus. Problem 5.2 shows that a function f from R2 to R is continuous, and defines the inverse image of an interval as the set of points mapping into that interval. Problem 5.3 provides a counterexample to show a function from R to R is not continuous by examining its inverse image of a closed set.

Uploaded by

Tom Davis
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

PMA307: Metric spaces Solutions 5

5.1. The denition is that f 1 [1, 2] = {(x, y ) R2 | f (x, y ) [1, 2]}. That is, it is the set of points {(x, y ) R2 | 1 with inner radius 1 and outer radius 2. x2 + y 2 2}, an annulus i.e. a ring,

5.2. (a) Let (x, y ) R2 . Take a sequence (xn , yn ) in R2 tending to (x, y ). Then (see the lecture notes!) xn x and yn y . By the algebra of limits, xn yn xy , or, in other words, f (xn , yn ) f (x, y ). Therefore f is continuous at (x, y ), and therefore (since (x, y ) was arbitrary) at all points of R2 . (b) By denition, f 1 [1, ) consists of all the points (x, y ) in R2 with f (x, y ) [1, ), i.e., with xy 1. This is exactly the set appearing in the question. As [1, ) is closed in R, and f is continuous, f 1 [1, ) is closed in R2 . 5.3. Let A = {0} (any closed subset of R containing 0 but not containing 1 will do). Then f 1 (A) consists of all real numbers x such that f (x) A, i.e., f (x) = 0. This set is exactly R \ {0}, which is not closed in R. Consequently, f is not continuous. 5.4. Let xn x in X . We need to show that f (xn ) f (x) in Y . This follows since dY (f (xn ), f (x)) dX (xn , x) 0 = 0.

5.5. We shall show that the inverse image of a closed subset of Y under h is a closed subset of X . Let F Y be closed. Then from the denition of h we see that h1 (F ) = f 1 (F ) g 1 (F ). Since f , g are continuous (from A, B to Y resp.), we see that f 1 (F ), g 1 (F ) are closed subsets of A, B respectively. Since A, B are closed, we can deduce that f 1 (F ), g 1 (F ) are in fact closed subsets of X . Hence h1 (F ) is closed, and so h is continuous. 5.6*. All the three statements are false in general. (a) Take X = (0, 1), Y = R with standard metrics and take f to be f (x) = x. Then f is continuous, X is closed in X but the image f (X ) = (0, 1) certainly isnt closed in R. (b) Take X = [0, 1], Y = R with standard metrics and take f to be f (x) = x. Then f is continuous, X is open in X but the image f (X ) = [0, 1] certainly isnt open in R. (c) Take X to be R with the discrete metric, and take Y to be R with the standard metric. Consider the function f : X Y given by f (x) = x. Clearly f is a bijection and f 1 also sends x to x. To see that f is continuous, take xn x in X . Since the metric is discrete, we must have xn = x for almost all n. Such a sequence clearly converges in Y . However, consider the sequence (1/n). In Y , we know 1/n 0. Under f 1 , we get the sequence 1/n in X and it doesnt converge! (cf. problem C2.) 5.7*. (a) Immediate from the denition since (an , bn ) (a, b) if and only if A (an , bn ) = an a = A (a, b) and B (an , bn ) = bn b = B (an , bn ). 1

(b) f :X AB is continuous

f (xn ) f (x) whenever xn x in X A (f (xn )) A (f (x)) and A (f (xn )) A (f (x)) whenever xn x in X A f and b f are continuous. 5.8*. (a) Let xn x in X . We want to show that dF (xn ) dF (x) in R. For any y F , the triangle inequality gives us that d(xn , y ) d(xn , x) + d(x, y ). On taking the inmum over all y F , we see that
y F

inf {d(xn , y )}

d(xn , x) + inf {d(xn , y )},


y F

and so dF (xn ) dF (x) Thus

d(xn , x). Similarly, we derive that dF (x) dF (xn ) d(xn , x) dF (xn ) dF (x) d(xn , x)

d(x, xn ).

and on taking limits as n we get dF (xn ) dF (x). Since d is always non-negative, we see that dF (x) 0. If x F , then clearly dF (x) = 0. We claim that the converse is also true: ie if dF (x) = 0 then x F . In this case, 1/n is not a lower bound of {d(x, y ) : y F } and so we can nd yn F with d(x, yn ) < 1/n. But then yn x as d(x, yn ) 0 and therefore x F as F is closed. (b) Since f is continuous and (, 0), (0, ) R are open, we get that U := f 1 (, 0), V := f 1 (0, ) are open subsets of X . It is easy to see that U V = (for if x U V then f (x) (, 0) (0, )). We now show that E U . If x E then dE (x) = 0. Since E F = , we see that x F and consequently by the previous part dF (x) > 0. Hence f (x) = 0 dF (x) < 0 and so x f 1 (, 0) = U , which then gives E U . A similar argument shows that F V .

PMA307

Semester 1, 2008-09

You might also like