Notes On The Chain Rule
Notes On The Chain Rule
h(x) = sin(x2 )
then
h(x) = f (g(x))
where
f (x) = and g(x) = .
So, can we find
d
f (g(x))
dx
knowing f 0 (x) and g 0 (x)?
In order to come up with the formula, we will change our notation. Let
y = f (g(x))
and
u = g(x)
then
y = f (u).
We want to compute
dy
dx
We remember that this was a rate of change (or a slope):
dy ∆y
= lim
dx ∆x→0 ∆x
∆y ∆u
= lim ·
∆x→0 ∆x ∆u
dy du
= ·
du dx
1
The First Version of The Chain Rule
dy dy du
= ·
dx du dx
This is easy to remember because it works like fraction arithmetic. You can apply it to any variables. We
use this version mainly in story problems where x, y and u are quantities represented possibly by different
letters.
to get the
Example 1:
d
sin x2 by first identifying the two functions making up the composition.
Use the Chain Rule to compute
dx
Example 2:
d 8
Use the Chain Rule to compute sin x − x3
dx
Example 3:
2
Use the Chain Rule to differentiate f (x) = ex
2
Example 4:
p
Differentiate f (x) = 1 + 5x2
Example 5:
2
Differentiate f (x) = esin(3x +6)
Example 5:
p
Differentiate f (x) = 1 + tan (ex + 5x)
Example 6:
p
Differentiate f (x) = sin cos(ex )
Example 7:
13
x3 + e5x
Differentiate f (x) = √
x sin x + ln x
3
The Second and Higher Derivatives
We start with f (x), differentiate to get f 0 (x) If we differentiate f 0 (x), we get f 00 (x).
Example 8:
Compute f 00 (x) if f (x) = x3 + e7x .
Notation:
f, f 0 , f 00 , f 000 , f iv , f v ,
After a while Roman numerals gets tricky so we switch to regular numbers. In order to differentiate (in the
English sense as in tell apart :-) ) powers of f from derivatives of f , we use parentheses:
We use third or higher derivatives in Math 126. They are here mainly for notation and computation.
We also have the quotient notation. The second derivative id the derivative of the first derivative:
d dy
=
dx dx
so we have
dy d2 y d3 y
, , , ...
dx dx2 dx3
Example 9:
Compute f (2021) (x) if f (x) = x3 + e7x + cos x.
ax+h − ax ah − 1
lim = ax lim
h→0 h h→0 h
eh − 1 d x
where e was the special number with lim = 1 so e = ex . Now, we will differentiate f (x) = ax
h→0 h dx
using this, the chain rule and a property of exponentials and logarithms. So, before we start, here is a review
of laws of exponentials and logarithms. Most must be familiar, some we will use again in the future.
4
LAWS OF EXPONENTIALS AND LOGARITHMS
1. ex = y is the same as saying ln y = x. In other words f (x) = ex and g(x) = ln x are inverse functions.
Therefore,
ln (ex ) = x and eln x = x.
In particular, ln e = 1.
2. ln(AB) = ln A + ln B.
This follows from the law of exponentials : ex · ey = ex+y .
A
3. ln = ln A − ln B.
B
ex
This follows from the law of exponentials : y = ex−y .
e
B
4. ln A = B · ln A.
y
This follows from the law of exponentials : (ex ) = exy .
ln B
5. logA B = .
ln A
Therefore,
d x
a = ln a · ax
dx
which also tells us
ah −1
limh→0 h = ln a.
Example 10:
d tan x
3
dx