Saturday, November 6, 2010

HW7 Solutions

Solutions to most problems are given. Please ask any questions which come up. Please let me know about any errors you find. You can leave a comment or send an email.

Homework 7:
3.10: 4, 14, 18, 20, 23, 28, 30, 32 a
4.1: 4, 34, 42, 53, 56, 68 b, 77
4.2: 14, 16, 18, 20, 24, 28, 32


3.10:

\#4

$f^{\prime}(x)=\frac{3}{4}x^{-1/4}$

$L(x)=f(1)+f^{\prime}(1)(x-1)=0+1(x-1)=x-1$

\#14a

Let $y=f(t)$.

Then $\frac{dy}{dt}=f^{\prime}(t)=e^{\tan\pi t}\cdot\sec^{2}(\pi t)\cdot\pi$.

Then $dy=\pi\sec^{2}(\pi t)e^{\tan\pi t}dt$

\#14b

$dy=\frac{1}{2z\sqrt{1+\ln z}}dz$

\#18a

$dy=-\sin xdx$

\#18b

$dy=-\sin\left(\frac{\pi}{3}\right)(0.05)=-\frac{\sqrt{3}}{2}(0.05)=-0.025\sqrt{3}$

\#20

$\Delta y=f(2)-f(1)=\sqrt{2}-\sqrt{1}\approx0.414$

$dy=\frac{1}{2\sqrt{x}}dx=\frac{1}{2}(1)=0.5$

\#23

We find the linearization of $f(x)=x^{5}$ at $a=2$. Then for values
$x$ near 2, we have $f(x)\approx L(x)$.

$f^{\prime}(x)=5x^{4}$, $f(2)=32$, $f^{\prime}(2)=80$

$L(x)=32+80(x-2)$

$(2.001)^{5}\approx L(2.001)=32+80(2.001-2)=32+80(.001)=32+.08=32.08$

Note that computing $(2.001)^{5}$ on the calculator gives approximately
$32.08008$.

\#28

We find the linearization of $f(x)=\sqrt{x}$ at $a=100$.

$f^{\prime}(x)=\frac{1}{2}x^{-1/2}$, $f(100)=10$, $f^{\prime}(100)=\frac{1}{20}$

$L(x)=10+\frac{1}{20}(x-100)$

$\sqrt{99.8}\approx L(99.8)=10+\frac{1}{20}(99.8-100)=10+\frac{1}{20}(-0.2)=10+(-0.01)=9.99$.

Note that computing $\sqrt{99.8}$ on the calculator gives approximately
$9.98999499$.

\#30

$f^{\prime}(x)=6x^{5}$

$L(x)=1+6(x-1)$

$L(1.01)=1.06$

$1.01$ is reasonably close to $1$ and so the approximation is reasonable.

\#32a

The linearizations of all three functions is the same $L(x)=1-2x$.
This is because they have the same function values and derivative
values at $a=0$.

4.1:

\#4

a: absolute minimum

b: neither

c: neither

d: local minimum

r: absolute maximum

s: local minimum

\#34

Note: In class I had the incorrect notion of critical numbers. See
page 274 for the book's definition.

The derivative does not exist at $t=\frac{4}{3}$

There are no points at which the derivative equals 0.

Thus the set of critical numbers is $\{\frac{4}{3}\}$

\#42

Note: In class I had the incorrect notion of critical numbers. See
page 274 for the book's definition.

$f(\theta)=4\theta-\tan\theta$ and it's $f^{\prime}(\theta)=4-\sec^{2}\theta$
exists everywhere on the domain of $f$.

We are left to find points at which the derivative equals 0.

$f^{\prime}(\theta)=0$ implies $\sec^{2}\theta=4$ implies $\sec\theta=\pm2$
implies $\cos\theta=\pm\frac{1}{2}$ implies $\theta=\frac{\pi}{3}+\pi k$
or $\frac{2\pi}{3}+\pi k$ for $k\in\mathbb{Z}$.

One could also write $\theta=\frac{\pi}{3}+2\pi k$, $\frac{2\pi}{3}+2\pi k$,
$\frac{4\pi}{3}+2\pi k$, $\frac{5\pi}{3}+2\pi k$ for $k\in\mathbb{Z}$.

\#53

Use the closed interval method on page 275.

\#56

Use the closed interval method on page 275.

\#68b

$f^{\prime}(x)$ exists on the domain of $f$ which is $-2\leq x\leq0$
and $f^{\prime}(x)=1+2\sin x$.

So we find numbers where $f^{\prime}(x)=0$. Thus we solve $\sin x=-\frac{1}{2}$.

In general we would have $x=-\frac{\pi}{6}+2\pi k$ , $-\frac{5\pi}{6}+2\pi k$
for $k\in\mathbb{Z}$, but our domain is $-2\leq x\leq0$.

Thus we only have $x=-\frac{\pi}{6}$ and $x=-\frac{5\pi}{6}$.

The values of $f$ at these points are $-\frac{\pi}{6}-2\frac{\sqrt{3}}{2}=-\frac{\pi}{6}-\sqrt{3}$
and $-\frac{5\pi}{6}-2\left(-\frac{\sqrt{3}}{2}\right)=-\frac{5\pi}{6}+2\sqrt{3}$

Now we compute $f$ at the endpoints. $f(0)=0-2(1)=-2$. $f(-2)=-2-2\cos(-2)$

Use a calculator to see which value is the absolute maximum and which
is the absolute minimum.

Note, had \#68a been assigned, one could see from the graph that the
values we'd want to compute is the maximum at $x=-2$ and the minimum
at $x=-\frac{\pi}{6}$.

\#77

Read the proof on page 273. The end of the proof tells you how to
prove Fermat's Theorem for the case of a local minimum.

4.2:

\#14

$f$ is continuous on the closed interval {[}1,4{]}

$f^{\prime}$ exists on the open interval $(1,4)$.

Thus the conditions hold.

$f^{\prime}(x)=\frac{(x+2)-x}{(x+2)^{2}}=\frac{2}{(x+2)^{2}}$

Then $f^{\prime}(c)=\frac{f(4)-f(1)}{4-1}=\frac{\frac{2}{3}-\frac{1}{3}}{3}=\frac{1}{9}$

Then $(x+2)^{2}=18$

$(x+2)=\pm3\sqrt{2}$

$x=-2\pm3\sqrt{2}$.

But only $x=-2+3\sqrt{2}\in(1,4)$.

\#16

It doesn't contradict the theorem because $f^{\prime}(x)$ does not
exist at $x=\frac{1}{2}\in(0,3)$.

\#18

One way to do this problem is to suppose it had more than one. Then
$f(x_{1})=f(x_{2})=0$ for two distinct numbers $x_{1}$, $x_{2}\in\mathbb{R}$.
Let's say $x_{1}<x_{2}$.

Because $f(x)=2x-1-\sin x$ is differentiable on $\mathbb{R},$ we
use the mean value theorem to imply that there exists a $c\in(x_{1},x_{2})$
such that $f^{\prime}(c)=\frac{f(x_{2})-f(x_{1})}{x_{2}-x_{1}}=0$.

But $f^{\prime}(x)=2-\cos x>0$. (There are various ways to argue
and/or see that $2-\cos x>0$, let me know if you don't know one way
to reason this)

Thus, we've reached a contradiction, implying that there is at most
one zero. Use the intermediate value theorem to obtain that $f$ has
at least one zero.

\#20

Approach the problem like the solution to \#18. Suppose there were
at least three. Then suppose $x_{1} values where the derivative equals $0$. But then the derivative $4x^{3}+4$
only has one zero. Contradiction. We conclude there are at most two
zeros.

In general, the same argument shows that for $f$ differentiable on
$\mathbb{R}$, if $f^{\prime}$has $n$ distinct real roots, then
$f$ has at most $n+1$ distinct real roots.

\#24

One can imitate Example 5 on page 284, or prove the exercise by contradiction
which I showed in section.

\#28

Apply the mean value theorem on the interval $[-b,b]$.

\#32

As the exercise indicates, follow example 6.

No comments:

Post a Comment