Scribd
Upload
307 views3 pages

Chain Rule

The document proves the chain rule for derivatives. It does so by: 1) Defining the derivative of the composite function f(g(x)) and introducing variables v and w for the remainders of Taylor series approximations; 2) Using these definitions and approximations to rewrite f(g(x+h)) in terms of derivatives and remainders; 3) Taking the limit as h approaches 0 to show the derivative is the product of the derivative of the outer function and the derivative of the inner function. The document also notes that many calculus texts present an incorrect "flawed" proof of the chain rule that assumes properties it does not prove.

Uploaded by

vinaroe
Copyright
© © All Rights Reserved
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
307 views3 pages

Chain Rule

The document proves the chain rule for derivatives. It does so by: 1) Defining the derivative of the composite function f(g(x)) and introducing variables v and w for the remainders of Taylor series approximations; 2) Using these definitions and approximations to rewrite f(g(x+h)) in terms of derivatives and remainders; 3) Taking the limit as h approaches 0 to show the derivative is the product of the derivative of the outer function and the derivative of the inner function. The document also notes that many calculus texts present an incorrect "flawed" proof of the chain rule that assumes properties it does not prove.

Uploaded by

vinaroe
Copyright
© © All Rights Reserved
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/ 3

Proof of the Chain Rule

Given two functions f and g where g is differentiable at the point x


and f is differentiable at the point g(x) = y, we want to compute the
derivative of the composite function f (g(x)) at the point x. In other
words, we want to compute
f (g(x + h)) f (g(x))
.
h0
h
lim

The answer involves the derivative of the outer function f and the
derivative of the inner function g. Specifically,
d(f g)
= (f g)0 (x) = (f (g(x))0 = f 0 (g(x)) g 0 (x).
dx
To prove this rule we first set up some notation. We are assuming
that the function g(x) is differentiable at the point x. This means
that the number g 0 (x) exists and is equal to our limit definition of the
derivative, and so
g(x + h) g(x)
g 0 (x) 0 as h 0.
h
We define a new variable v by
v=

g(x + h) g(x)
g 0 (x).
h

Notice that v depends on the number h and that v 0 as h 0.


Similarly, because we are assuming the function f is differentiable at
the point y = g(x) we have
f (y + k) f (y)
f 0 (y) 0 as k 0,
k

where here we are using the variable k to stand for the h we normally
use. Again, we define another variable w by
f (y + k) f (y)
w=
f 0 (y).
k
Again, notice that w depends on k and that w 0 as k 0.
From our definitions of v and w it follows that
g(x + h) = g(x) + [g 0 (x) + v]h
f (y + k) = f (y) + [f 0 (y) + w]k
We now use these equations to rewrite f (g(x + h)). In particular, use
the first equation to obtain
f (g(x + h)) = f (g(x) + [g 0 (x) + v]h),
and use the second equation applied to the right-hand-side with k =
[g 0 (x) + v]h and y = g(x). Note that using this quantity for k tells us
that k 0 as h 0, and so w 0 as h 0. Applying this equation
gives
f (g(x) + [g 0 (x) + v]h) = f (g(x)) + [f 0 (g(x)) + w] [g 0 (x) + v]h.
Using this last equation we now simplify the fraction
f (g(x + h)) f (g(x)) f (g(x)) + [f 0 (g(x)) + w] [g 0 (x) + v]h f (g(x))
=
h
h
[f 0 (g(x)) + w] [g 0 (x) + v]h
=
h
= [f 0 (g(x)) + w] [g 0 (x) + v].
2

We are now ready to compute the derivative:


f (g(x + h)) f (g(x))
= lim [f 0 (g(x)) + w] [g 0 (x) + v]
h0
h0
h
lim

= lim f (g(x)) + lim w


h0



h0

lim g (x) + lim v

h0

h0

= f 0 (g(x)) g 0 (x),
since v 0 as h 0 and w 0 as h 0.
The Flawed Proof Many calculus texts present an incorrect proof
of the chain rule that goes as follows:
f (g(x + h)) f (g(x))
h0
h

(f g)0 (x) = lim

(f g) (x)

1
g 0 (x)

f (g(x + h)) f (g(x))


h

 

f (g(x + h)) f (g(x))


g(x + h) g(x)

= lim

h0

= lim

h0

h
g(x + h) g(x)

= f 0 (g(x)).

Therefore (f g)0 (x) = f 0 (g(x)) g 0 (x).


For 3 bonus points, explain why this proof is technically incorrect.

You might also like