CHAPTER 2: Partial Derivatives

2.1 Definition of a Partial Derivative

2.2 Increments and Differential

2.3 Chain Rules

2.4 Local Extrema

2.1 Definition of a Partial Derivative

Definition
If z  f ( x, y ) , then the (first) partial
derivatives of f with respect to x and y are the
functions f x and f y respectively defined by

f ( x   x, y )  f ( x, y )
f x  lim
x  0 x

f ( x , y   y )  f ( x, y )
f y  lim
y  0 y

provided the limits exist.

2.1.1 Notation
For z  f ( x, y ) , the partial derivatives f x
and f y are also denoted by the symbols:
 f z 
, , f ( x, y ) , f x ( x, y ) or z x
 x  x x

f z 
, , f ( x, y ) , f y ( x, y ) or z y
 y  y y
The values of the partial derivatives at the point
(a, b) are denoted by

f f
 f x (a, b) and  f y ( a, b)
x ( a,b) y ( a,b)

Note
 The stylized “d” symbol in the notation is
called roundback d, curly d or del d.
 It is not the usual derivative d (dee) or 
(delta d).
Example
If f (x ,y ) x 3y x 2y 2 , find
f f
i. ii. iii. fy (1, 1)
x y

Solution

(a) For f x , hold y constant and find the

derivative with respect to x:
f
x 3y x 2y 2 3x 2y 2xy 2
x x

(b) For f y , hold x constant and find the

derivative with respect to y:
f
x 3y x 2y 2 x3 2x 2y
y y

(c) fy (1, 1) (1)3 2(1)2 ( 1) 2

For a function f ( x, y, z ) of three variables,
there are three partial derivatives:
f x , f y and f z
The partial derivative f x is calculated by
holding y and z constant. Likewise, for f y and
fz.

2.1.2 Partial Derivative as a Slope

Example 2.6
Find the slope of the line that is parallel to the
xz-plane and tangent to the surface
z  x x  y at the point P(1, 3, 2) .

Solution

Given f ( x, y )  x x  y

WANT: f x (1, 3)
  1
12
f x ( x, y )  ( x  y )  x  ( x  y ) 1 2 (1  0)
 2

x
 x y
2 x y

Thus the required slope,

1 9
f x (1, 3)  1  3  
2 1 3 4

2.1.3 Partial Derivative as a Rate of

Change
 A partial derivative is the rate of change of
a multi-variable function when we allow
only one of the variables to change.
f
 Specifically, the partial derivative at
x
( x0 , y0 ) gives the rate of change of f with
 respect to x when y is held fixed at the
value y0.

Example
The volume of a gas is related to its temperature
T and its pressure P by the gas law PV  10T ,
where V is measured in cubic inches, P in
pounds per square inch, and T in degrees
Celsius. If T is kept constant at 200, what is the
rate of change of pressure with respect to
volume at V  50?

Solution
P
WANT:
V T  200, V  50

Given PV  10T .

P  10T

V V2
 P (10)(200) 4
  
V T  200, V 50 (50) 2
5

2.1.4 Higher Order Partial Derivatives

Standard Notations
Given z  f ( x, y )

Second partial derivatives

 2 f   f 
=   = ( f x ) x = f xx
x 2 x  x 

 2 f   f 
=   = ( f y ) y = f yy
y 2 y   y 
Mixed partial derivatives

2 f   f 
=   = ( f y ) x = f yx
xy x  y 

2
 f   f 
=   = ( f x ) y = f xy
yx y  x 

3 f     f  
=   = f yyx
xy 2  
 x  y  y  

4 f       f   
=       = f yyxx
2
x y 2  x   x  y   y   
Question 1
Find the first partial derivative of f (x, y ).
2
2 3
i. f x , y x y
ii. f x, y e x ln xy
Question 2
Find the first partial derivative of
f x, y, z cos xy sin yz .
Question 3
Find the value of fxy (1, 0), fyx (1, 0), fxx (1, 0) and
fyy (1, 0)for the given

f x, y x y sin x y .
2.2 Increments and Differential
Let z  f ( x, y ) , where x and y are
independent variables. If x is subject to a small
increment (or a small error) of x, while y
remains constant, then the corresponding
increment of z in z will be
z
z  x
x
Similarly, if y is subject to a small increment of y,
while x remains constant, then the corresponding
increment of z in z will be
z
z  y
y
It can be shown that, for increments (or errors)
in both x and y,
z z
z  x  y
x y
The formula for a function of two variables may
be extended to functions of a greater number of
independent variables.
For example, if w  f ( x, y, z ) of three
variables, then
w w w
w  x  y  z
x y z

Definition
Let z  f ( x, y ) where f is a differentiable
function and let dx and dy be independent
variables. The differential of the dependent
variable, dz is called the total differential of z is
defined as
dz  df ( x, y)  f x ( x, y)dx  f y ( x, y)dy
Thus, z  dz provided dx is the change in x
and dy is the change in y.
Example
3 3
Let f ( x, y )  2 x  xy  y . Compute z
and dz as (x, y) changes from (2, 1) to
(2.03, 0.98).

Solution
z = f(2.03, 0.98) f(2, 1)

 2(2.03)3  (2.03)(0.98)  (0.98)3

[2(2) 2  2(1)  13 ]
= 0.779062
dz  f x ( x, y)dx  f y ( x, y)dy

 (6 x 2  y )x  ( x  3 y 2 )y

At (2, 1) with x = 0.03 and y = 0.02,

dz  (25)(0.03)  (1)(0.02)  0.77

Example
Suppose that a cylindrical can is designed to have
a radius of 1 in. and a height of 5 in. but that the
radius and height are off by the amounts dr =
0.03 and dh = 0.1. Estimate the resulting
absolute, relative and percentage changes in the
volume of the can.
Solution
WANT: Absolute change, V dV
V dV
Relative change,
V V
dV
Percentage change,  100
V
Absolute change,

dV  Vr dr  Vh dh  2rhdr  r 2 dh

 2 (1)(5)(0.03)   (1) 2 (0.1)  0.2

Relative change,

dV 0.2 0.2
 2   0.04
V 2
r h  (1) (5)

Percentage change,

dV
 100  0.04  100  4%
V
Question 1

Use partial derivatives to find an approximate

change in

f ( x, y, z )  x 2 z 3  3 yz 2  x 3  2 yz
if ( x, y, z ) changes from (1,4,2) to
(1.02,3.97,1.96) .

Question 2
The radius and height of a right circular cone
are measured with errors of at most 3% and 2%
respectively. Use differentials to estimate the
maximum percentage error in computing the
volume.
2.3 Chain Rule

2.3.1 Partial Derivatives of Composite

Functions
If y is a differentiable function of x and x is a
differentiable function of a parameter t, then
the chain rule states that
dy dy dx
 
dt dx dt
Theorem 1
If z  f ( x, y ) is differentiable and x and y are
differentiable functions of t, then z is a
differentiable function of t and
dz z dx z dy
   
dt x dt y dt
Theorem 2
Let x  x(r , s) and y  y (r , s) have partial
derivatives at r and s and let z  f ( x, y ) be
differentiable at (x, y). Then
z  f ( x(r , s), y(r , s)) has first derivatives
given by

z z x z y
   
r x r y r
z z x z y
   
s x s y s
Question 1

Use the chain rule to find z x and z y .

i. z uv uv 2, u x sin y , v y sin x
ii. z r3 s v 2, r xe y , s ye x , v x 2y

Question 2

Use the chain rule to find w t .

2t
w ln u v ,u e ,v t3 t2

Question 3

The radius r and the altitude h of a right circular

cylinder are increasing at the rates of 0.01
cm/min and 0.02 cm/min, respectively. Use the
chain rule to find the rate at which the volume is
increasing at the time when r  8 cm and
h  14cm. At what rate is the curved surface
area increasing at this time?
2.3.2 Partial Derivatives of Implicit
Functions
The chain rule can be applied to implicit
relationships of the form F ( x, y )  0 .

Differentiating F ( x, y )  0 with respect to x

gives

F dx F dy
    0.
x dx y dx
F F dy
In other words,    0.
x y dx
dy  F x
Hence,  .
dx F  y
In summary, we have the following results.
Theorem 3
If F ( x, y )  0 defines y implicitly as a
differentiable function of x, then

dy  Fx  x, y 
 .
dx Fy  x, y 
Theorem 4
If F ( x, y, z )  0 defines z implicitly as a
differentiable function of x and y, then

z  Fx  x, y, z  z  Fy  x, y, z 
 and  .
x Fz  x, y, z  y Fz  x, y, z 
Question 1

In equations 1(i) and 1(ii), the equation implicitly

defines a function of one variable y as a function
of x. Use Theorem 3 to find y x .
i. 2x 3 xy 2 y 3 1
ii. x xy 3y 4

Question 2
In equations 2(i) and 2(ii), the equation implicitly
defines a function of two variables z as a
function of x and y. Use Theorem 4 to find z x
and z y .
i. xeyz 2ye xz 3ze xy
ii. x y 2 z sin z
2.5 Local Extrema

Definition 1
A function of two variables has a local
maximum at (a, b) if f (x,y ) f (a,b) when (x, y)
is near (a, b). The number f (a,b) is called a
local maximum value.
If f (x,y ) f (a,b) when (x, y) is near (a,b), then
f (a,b) is a local minimum value.

Theorem 1
If f has a local maximum or minimum at (a, b)
and the first-order partial derivatives of f exist
at this point, then fx (a,b) 0 and fy (a,b) 0 .

Definition 2
A point (a, b) is called a critical point of the
function z f (x,y ) if fx (a,b) 0 and fy (a,b) 0
or if one or both partial derivatives do not exist
at (a, b).
Relative Max

Point (a, b, f (a, b)) is a local maximum

Relative Min.
Point (a, b, f (a, b)) is a local minimum

Saddle Point

Point (a, b, f (a, b)) is a saddle point

Remark
The values of z at the local maxima and local
minima of the function z f (x,y ) may also be
called the extreme values of the function,
f (x,y ).
Theorem 2 : Second-Partials Test
Let f (x,y ) have a critical point at (a, b) and
assume that f has continuous second-order
partial derivatives in a disk centered at (a, b).
Let
D fxx (a,b)fyy (a,b) [ fxy (a,b)]2

(i) If D > 0 and fxx (a,b) 0 , then f has a local

minimum at (a, b).
(ii) If D > 0 and fxx (a,b) 0 , then f has a local
maximum at (a, b).
(iii) If D < 0, then f has a saddle point at (a, b).
(iv) If D = 0, then no conclusion can be drawn.
Remark
The expression fxx fyy fxy 2 is called the
discriminant or Hessian of f. It is sometimes
easier to remember it in the determinant form,
fxx fxy
fxx fyy fxy 2
fxy fyy

If the discriminant is positive at the point (a, b),

then the surface curves the same way in all
directions:
 downwards if fxx 0 , giving rise to a local
maximum
 upwards if fxx (a,b) 0 , giving a local
minimum.
If the discriminant is negative at (a, b), then the
surface curves up in some directions and down
in others, so we have a saddle point.
Example
Locate all local extrema and saddle points of
f (x,y ) 1 x 2 y 2 .

Solution
 First determine fx and fy:
fx (x,y ) 2x and fy (x,y ) 2y .

 Secondly, solve the equations,

fx = 0 and fy = 0 for x and y:
2x 0 and 2y 0
So the only critical point is at (0, 0).
 Thirdly, evaluate fxx , fyy and fxy at the critical
point.
fxx (x,y ) 2 , fxy (x,y ) 0 and fyy (x,y ) 2

At the point (0, 0),

 fxx (0,0) 2 , fxy (0,0) 0 and
fyy (0,0) 2

 Compute D:
2 0
D 4
0 2

Since D = 4 > 0 and f xx (0, 0)  2 < 0, the

second partials test tell us that a local
maximum occurs at (0, 0).
In other words, the point (0, 0, 1) is a local
maximum, with f having a corresponding
maximum value of 1.
Question

In equations 1 - 3 , find critical points of f  x, y 

and determine whether f  x, y  at that point is
a local maximum or a local minimum, or the
value of the saddle point.
1. f x, y 3 x 2 y 2 6y
2. f x, y x2 y2 4x 6y 23
3. f x, y x2 y2 2x 6y 4