Homework 7

Table of Contents

Exercise 17.6

A rational function is a function ff of the form pq\frac{p}{q} where pp and qq are polynomial functions. The domain of ff is {xRq(x)0}\set{x \in \R \mid q\p{x} \neq 0}. Prove every rational function is continuous. Hint: Use Exercise 17.5.

Solution.

By Exercise 17.5, polynomials are continuous functions on R\R. (This is not hard to prove--you can prove that xnx^n is continuous for each nNn \in \N by induction; then every polynomial is a finite linear combination of these.) Thus, by Theorem 17.4(iii), pq\frac{p}{q} is continuous at x0x_0 provided q(x0)0q\p{x_0} \neq 0, which is exactly the domain of ff.

Exercise 18.4

Let SRS \subseteq \R and suppose there exists a sequence (xn)\p{x_n} in SS converging to a number x0Sx_0 \notin S. Show there exists an unbounded continuous function on SS.

Solution.

Consider f(x)=1xx0f\p{x} = \frac{1}{x - x_0}. Since x0Sx_0 \notin S, it follows that xx00x - x_0 \neq 0 for all xSx \in S. In particular, by Theorem 17.4(iii) again, ff is a continuous function on SS.

To show that ff is unbounded, let M>0M > 0. Since xnx0x_n \to x_0, there exists nNn \in \N such that xnx0<1M\abs{x_n - x_0} < \frac{1}{M}, so for this nn, we have

f(xn)=1xnx0>M.\abs{f\p{x_n}} = \frac{1}{\abs{x_n - x_0}} > M.

Since MM was arbitrary, this means ff is unbounded.

Exercise 18.10

Suppose ff is continuous on [0,2]\br{0, 2} and f(0)=f(2)f\p{0} = f\p{2}. Prove there exist x,y[0,2]x, y \in \br{0, 2} such that yx=1\abs{y - x} = 1 and f(x)=f(y)f\p{x} = f\p{y}. Hint: Consider g(x)=f(x+1)f(x)g\p{x} = f\p{x + 1} - f\p{x} on [0,1]\br{0, 1}.

Solution.

We follow the hint. Note that xf(x+1)x \mapsto f\p{x + 1} is continuous since it is the composition of two continuous functions, so gg itself is continuous since it is the difference of two continuous functions.

By definition, we have

g(0)=f(1)f(0)=(f(0)f(1))=g(1).g\p{0} = f\p{1} - f\p{0} = -\p{f\p{0} - f\p{1}} = -g\p{1}.

Thus, 00 lies between (or is one of) g(0)g\p{0} and g(1)g\p{1}. Thus, by the intermediate value theorem, there exists x0[0,1]x_0 \in \br{0, 1} such that g(x0)=0g\p{x_0} = 0. But this is equivalent to

f(x0+1)=f(x0),f\p{x_0 + 1} = f\p{x_0},

so we are done by setting (x,y)=(x0,x0+1)\p{x, y} = \p{x_0, x_0 + 1}.