Differential Calculus of Functions of Several Variables
Differential Calculus of Functions of Several Variables
Definition 5.1.1
each element in D.
The set D is the domain of f . The set of w values taken on by f is the function's range.
Example 5.1.1
xy 5 1
(a) f ( x, y) (b) f ( x, y ) sin ( xy ) (c) f ( x, y, z )
2 y x2 x y2 z2
2
Solutions:
(a) The domain D is the set of all pairs ( x, y ) such that y x 2 0, i.e.
D {( x, y ) 2
: y x 2 }.
The range is the set of all real numbers.
(b) The domain is the entire plane 2
and the range is the interval [1,1].
(c) The domain D is the set of all pairs ( x, y, z ) such that x 2 y 2 z 2 (0, 0, 0), i.e.
D {( x, y, z ) 3
: x 2 y 2 z 2 (0, 0, 0)}.
Since the denominator is positive, the range is the interval (0, ).
Definition 5.1.2
{( x, y) : | ( x, y) ( x0 , y0 ) | r}.
1
(ii) The boundary of the open disk is the circle
{( x, y) : | ( x, y) ( x0 , y0 ) | r}.
Definition 5.1.2
lim f ( x, y ) L,
( x , y ) ( x0 , y0 )
| f ( x, y ) L |
Example 5.1.2
Proofs:
| f ( x, y ) L | | x 3 y 4 | | ( x 1) (3 y 3) | | ( x 1) 3( y 1) |
| x 1| 3 | y 1|, by Triangle inequality
Thus,
1
Choose so that
4
1
| f ( x, y ) L | 4 ( x 1) 2 ( y 1) 2 4 4
4
Hence proved!
2
(ii) Let 0 be given. If 0 ( x 0)2 ( y 0)2 , then
xy 2 | x | .y2
| f ( x, y ) L | 0
x2 y 2 x2 y 2
x 2 y 2 .( x 2 y 2 )
, since | x | x 2 y 2 and y 2 x 2 y 2
x y
2 2
x2 y 2
Choose so that
| f ( x, y) L | x2 y 2
whenever 0 x 2 y 2 .
Hence proved!
The following rule for nonexistence of a limit of a function of two variables can be applied to
evaluate a limit:
Example 5.1.3
x xy 3 x 2 xy xy x2 y xy 2
(a) lim (b) lim x y
(c) lim (d) lim (e) lim
( x , y ) (0,1) xy 5 xy y ( x , y ) (0,0) x y ( x , y ) (0,0) x y ( x , y ) (0,0) x y
2 2 2 2 4 2 2 2
( x , y ) (0,0)
Solutions:
x xy 3 0 0(1) 3
(a) lim 3
( x , y ) (0,1) xy 5 xy y 0(1 ) 5(0)(1) 12
2 2 2
x 2 xy
lim x y
( x , y ) (0,0)
3
(c) It can be shown that along the paths x 0 and y 0, the limit is 0. But along the path
m
lim xy
lim x ( mx )
.
( x , y ) (0,0) x y ( x , y ) (0,0) x ( mx ) 1 m2
2 2 2 2
lim xy
does not exist.
( x , y ) (0,0) x y
2 2
(d) Check that along the paths x 0, y 0 and y mx the limit is 0. But along the path
y x 2 , we get
x2 y x2 ( x2 ) 1
lim lim
( x , y ) (0,0) x y ( x , y ) (0,0) x ( x )
4 2 4 2 2
2
x2 y
lim does not exist.
( x , y ) (0,0) x y
4 2
(e) Check that along the paths x 0 and y 0 the limit is 0. Along the path y mx
xy 2 x ( mx )2
lim lim lim m 2 x3
lim m2 x
0, for all m .
( x , y ) (0,0) x y ( x , y ) (0,0) x ( mx ) ( x , y ) (0,0) m x ( x , y ) (0,0) m
2 2 2 2 2 2 2 2
x 1
xy 2
lim 0
( x , y ) (0,0) x y
2 2
Definition 5.1.3
1. f is defined at ( x0 , y0 )
2. lim f ( x, y ) exists
( x , y ) ( x0 , y0 )
3. lim f ( x, y) f ( x0 , y0 ).
( x , y ) ( x0 , y0 )
A function is continuous on S 2
if it is continuous at every point ( x, y ) in S.
4
Example 5.1.5
Determine the continuity of the following at the indicated point:
x2 y
, ( x, y ) (0, 0)
(a) f ( x, y ) x 4 y 2 at (0, 0)
0, ( x, y ) (0, 0)
xy 2
, ( x, y ) (0, 0)
(b) f ( x, y ) x 2 y 2 at (0, 0)
0, ( x, y ) (0, 0)
(c) f ( x, y ) 5 x3 y 1xy 9 37 x 23 y 5 x 47 6 on 2
.
Solutions:
x2 y
lim f (0, 0)
( x , y ) (0,0) x y
4 2
Since the last condition is not satisfied, we conclude that f is not continuous at (0, 0).
(b) Check that all the conditions for continuity are satisfied. Thus, f is continuous at
(0, 0).
(c) The function
f ( x, y ) 5 x3 y 1xy 9 37 x 23 y 5 x 47 6
2 2
is defined on , the limit exists at every point in and it is equal to the value of
2
f ( x, y ). Thus, f ( x, y ) is continuous on .
Definition 5.1.4
lim f ( X ) L,
X X0
| f (X ) L |
whenever 0 ( x1 ( x1 )0 )2 ( x2 ( x2 )0 ) 2 ... ( xn ( xn )0 ) 2 .
5
Example 5.1.5
1. Use the definition to prove that
lim ( x1 2 x2 x3 3x4 ) 7
( x1 , x2 , x3 , x4 ) ( 4,1,0,3)
x5 2 x4 2 x32 x2 2 x1
2. Show that if f ( x1 , x2 ,..., x5 ) , then lim f ( x1 , x2 ,..., x5 )
x12 x2 2 x32 x4 2 x5 2 ( x1 ,..., x5 ) (0,...,0)
| f ( X ) L | | x1 2 x2 x3 3 x4 7 | | ( x (4)) (2 x2 2) ( x3 0) (3 x 9) |
| ( x 4) 2( x2 1) ( x3 ) 3( x 3) |
| x 4 | 2 | x2 1| | x3 | 3 | x4 3 |
( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2
2 ( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2
( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2
3 ( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2
7 ( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2
1
Choose so that
7
1
| f ( X ) L | 7 ( x1 ( 4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2 7 7
7
whenever 0 ( x1 (4)) 2 ( x2 1) 2 ( x3 0) 2 ( x4 3) 2 .
Hence proved!
2. Choosing the path x1 0 x2 x3 x4 , we get
x5 2
lim f ( x1 , x2 ,..., x5 ) lim 1
( x1 ,..., x5 ) (0,...,0) ( x1 ,..., x5 ) (0,...,0) x 2
5
x4 2
lim f ( x1 , x2 ,..., x5 ) lim 1
( x1 ,..., x5 ) (0,...,0) ( x1 ,..., x5 ) (0,...,0) x 2
4
6
5.2. PARTIAL DERIVATIVES
Recall that if y f ( x) is a function of one independent variable, then the derivative is the rate
of change of the function. In a similar way, the derivative of a function of several variables
gives the rate of change of the function. Since there are many independent variables in this
case, we have to choose one variable as the independent variable and keep all others constant.
When we differentiate a function of several variables in this way, we obtain the “partial
derivative”. For example, if z f ( x, y ) is a function of two variables, and we choose x as the
independent variable, then the partial derivative of z with respect to x is written symbolically
as
z f
or or f x ( x, y ),
x x
and is read “the partial of z with respect to x " or “the partial of f with respect to x ".
Similarly, if y is chosen as the independent variable, then the partial derivative of z with
respect to y is written symbolically as
z f
or or f y ( x, y ).
y y
Definition 5.2.1
Let z f ( x, y ) be a function of two variables.
z f f ( x x, y ) f ( x, y )
lim .
x x x0 x
f
The partial derivative is defined at every point ( x, y ) in the domain of f such
x
that the limit above exists.
(ii) The partial derivative of f with respect to y is the function
z f f ( x, y y ) f ( x, y )
lim .
y y y 0 y
f
The partial derivative is defined at every point ( x, y ) in the domain of f such
y
7
We can extend Definition 5.2.1 to a case of more than two variables.
Example 5.2.1
U U
(a) (b) U w (v, w, x, y, z ) (c)
v z
Solutions:
wy 2 z 3 vw
U (v, w, x, y, z )
x
w 2 3
y z v sin ( wz 2
)
x
sin ( wz 2 )
x
U wy 2 z 3 vw wy 2 z 3 vw 2
(a) sin ( wz 2 ) sin ( wz )
v v x x v x v x
w
0 sin ( wz 2 )
x
w
sin ( wz 2 )
x
U wy 2 z 3 vw 2 wy 2 z 3 vw 2
(b) U w (v, w, x, y, z ) sin ( wz ) sin ( wz )
w w x x w x w x
y2 z3 v vwz 2
sin ( wz )
2
cos ( wz 2 )
x x x
U wy 2 z 3 vw wy 2 z 3 vw 2
(c) sin ( wz 2 ) sin ( wz )
z z x x z x z x
3wy 2 z 2 2vwz
cos ( wz 2 )
x x
U U
Exercise: Find and
x y
8
Definition 5.2.2
Example 5.2.2
1. The function
f ( x, y, z ) 2 x 2 y 3xyz z 3
is homogeneous of degree 3 because
f (tx, ty, tz ) 2(tx 2 )(ty ) 3(tx )(ty )(tz ) (tz 3 )
2t 3 x 2 y 3t 3 xyz t 3 z 3
t 3 (2 x 2 y 3xyz z 3 )
t 3 f ( x, y , z )
2. The function
x3 x 2 y xy 2 y 3
f ( x, y)
x2 y 2
is homogeneous of degree 1 because
(tx)3 (tx) 2 (ty ) (tx)(ty ) 2 (ty )3
f (tx, ty )
(tx) 2 (ty ) 2
t 3 x3 t 3 x 2 y t 3 xy 2 t 3 y 3
t 2 x2 t 2 y 2
t 3 ( x 3 x 2 y xy 2 y 3 )
t 2 ( x2 y 2 )
x3 x 2 y xy 2 y 3
t
x2 y 2
tf ( x, y )
x
i 1
f
i xi tf ( x1 , x2 ,..., xn )
n
for all ( x1 , x2 ,..., xn ) in .
9
Example 5.2.3
Verify the Euler’s theorem for the following homogeneous functions:
x3 x 2 y xy 2 y 3
(a) f ( x, y, z ) 2 x y 3xyz z
2 3
(b) f ( x, y)
x2 y 2
Solutions:
f f f
(a) f ( x, y, z ) 2 x 2 y 3xyz z 3 4 xy 3 yz, 2 x 2 3xz and 3xy 3z 2
x y z
Thus,
f f f
x y z x 4 xy 3 yz y 2 x 2 3xz z 3xy 3z 2
x x x
4 x 2 y 3 xyz 2 x 2 y 3xyz 3xyz 3z 3
6 x 2 y 9 xyz 3z 2
3 2 x 2 3 xyz z 3
3 f ( x, y , z )
x3 x 2 y xy 2 y 3
(b) If f ( x, y ) , then
x2 y 2
f ( x 2 y 2 )(3x 2 2 xy y 2 ) 2 x( x 3 x 2 y xy 2 y 3 ) x 4 2 x 2 y 2 y 4 x y
2 2 2
1
x x2 y 2 x2 y 2 x2 y 2
2 2 2
and
f ( x 2 y 2 )( x 2 2 xy 3 y 2 ) 2 y ( x 3 x 2 y xy 2 y 3 ) x 4 2 x 2 y 2 y 4 x y
2 2
2
1
y x 2
y
2 2
x 2
y
2 2
x 2
y
2 2
x3 x 2 y xy 2 y 3
Note that f ( x, y ) can also be written as
x2 y 2
x3 x 2 y xy 2 y 3 ( x3 x 2 y ) ( xy 2 y 3 ) ( x y )( x 2 y 2 )
f ( x, y) x y
x2 y 2 x2 y 2 x2 y 2
Thus,
f f
x y x y f ( x, y ).
x y
10
5.2.1. Higher-Order Partial Derivatives
2 f
f xx ( x, y)
x x x 2
2 f
f yx ( x, y )
x y xy
2 f
f xy ( x, y)
y x yx
2 f
f yy ( x, y )
y y y 2
2 f 3 f
f xxx ( x, y )
x x 2 x3
2 f 3 f
f yxx ( x, y )
x xy x 2y
and so on.
Definition 5.2.3
2 f 2 f
f yx ( x, y ) and f xy ( x, y )
xy yx
Theorem 5.2.2
11
Example 5.2.4
2U 3U 3U
1. (i) (ii) (iii) at (0,1,1, 1,1).
x 2 x 2 w vwz
2. Show that
Solutions:
wy 2 z 3 vw
From Example 5.2.1, U (v, w, x, y, z )
x
y z v sin ( wz 2 )
w 2 3
x
sin ( wz 2 )
x
U wy 2 z 3 vw wy 2 z 3 vw 2
1. (i) sin ( wz 2 ) sin ( wz )
x x x x x x x x
wy 2 z 3 vw
2
2 sin ( wz 2 )
x x
2U wy 2 z 3 vw 2 2
2 2 sin ( wz 2 ) wy 2 z 3 3 vw 3 sin ( wz 2 )
x 2
x x x x x
2wy 2 z 3 2vw
3
3 sin ( wz 2 )
x x
U y2 z3 v vwz 2
(ii) Since U ( v , w, x , y , z ) sin ( wz 2
) cos ( wz 2 ) , we have that
w
w
x x x
1 1 1
U wx (v, w, x, y, z ) y 2 z 3 2 vwz 2 cos ( wz 2 ) 2 v 2 sin ( wz 2 )
x x x
2 3 2
y z vwz v
2 2 cos ( wz 2 ) 2 sin ( wz 2 )
x x x
3U 2 y 2 z 3 2vwz 2 2v
U ( v , w, x , y , z ) cos ( wz 2 ) 3 sin ( wz 2 )
x w
2 wxx 3 3
x x x
3wy 2 z 2 2vwz
(iii) Since U z (v, w, x, y, z ) cos ( wz 2 ), we have that
x x
12
3 y 2 z 2 2vz
U zw (v, w, x, y, z ) 2 w cos ( wz 2 ) w2 z 2 sin ( wz 2 )
x x
2 2
3y z 4vwz 2vw2 z 3
cos ( wz 2 ) sin ( wz 2 )
x x x
4 wz 2 w2 z 3
U zwv (v, w, x, y, z ) 0 cos ( wz )
2
sin ( wz 2 )
x x
2 3
4 wz 2w z
cos ( wz 2 ) sin ( wz 2 )
x x
4(1).(1) 2(12.13 )
U zwv (0,1,1, 1,1) cos (1.1 )
2
sin (1.12 )
1 x
4cos1 2sin1
wy 2 z 3 vw
2. (i) U x (v, w, x, y, z ) 2 sin ( wz 2 )
x2 x
2wyz 3 2wyz 3
U xy (v, w, x, y, z ) 0
x2 x2
U wy 2 z 3 vw 2 2wyz 3 2wyz 3
sin ( wz ) 0
y y x y x x x
2wyz 3
U yx (v, w, x, y, z )
x2
U xy (v, w, x, y, z) U yx (v, w, x, y, z). Hence shown!
y2 z3 v vwz 2
(ii) Since U w (v, w, x, y, z ) sin ( wz 2 ) cos ( wz 2 ) , we have that
x x x
1 wz 2
U wv (v, w, x, y, z ) 0 sin ( wz 2 ) cos ( wz 2 )
x x
2
1 wz
sin ( wz 2 ) cos ( wz 2 )
x x
w
and Since U v (v, w, x, y, z ) sin ( wz 2 ) , we have that
x
U vw (v, w, x, y, z )
1
x
sin ( wz 2 ) wz 2 cos( wz 2 )
1 wz 2
sin ( wz )
2
cos ( wz 2 )
x x
U vw (v, w, x, y, z ) U wv (v, w, x, y, z ). Hence shown!
13
5.3. TANGENT PLANE AND NORMAL LINE TO A SURFACE
In this section, we use partial derivatives to find the equations of the tangent line, normal line
and tangent plane to a surface in space z f ( x, y ), which can also be written as
F ( x, y, z ) z f ( x, y) 0.
Example 5.3.1
Find the equation of the tangent line to the surface z x 2 12 y 2 at the point
5, 2,3 that
lies
z z
(a) z x 2 12 y 2 y 2
y y ( x0 , y0 ) 5 , 2
Using ( y, z ) (2,3) and the slope 2, we get
z 3 2( y 2).
the equation of the tangent line to the surface that lies in the plane x 5 is
y 2 z 3
x 5,
1 2
14
z z
(b) z x 2 12 y 2 2x 2 5
x x ( x0 , y0 ) 5 , 2
Using ( x, z )
5,3 and the slope 2 5, we get
z 3 2 5( x 5).
the equation of the tangent line to the surface that lies in the plane y 2 is
x 5 z 3
, y 2
1 2 5
We now turn our attention to the equation of the normal line at tangent plane.
It should be noted that not every surface has a tangent plane at every point. For example, a cone
below has no tangent plane at the origin:
Definition 5.3.1
Let F be a scalar equation of three variables such that Fx , Fy and Fz exist at a point
F ( X 0 ) Fx ( x0 , y0 , z0 ) i Fy ( x0 , y0 , z0 ) j Fz ( x0 , y0 , z0 ) k.
15
Example 5.3.2
Find F ( X 0 ) if F ( x, y, z ) x ln y z ln x at X 0 (1,1,1).
Solutions:
1
Fx ( x, y, z ) ln y z. Fx (1,1,1) 1
x
1
Fy ( x, y, z ) x. 0 Fy (1,1,1) 1
y
Fz ( x, y, z ) ln x Fz (1,1,1) 0
F ( X 0 ) i j 0 k
Now, for a given surface F ( x, y, z ) z f ( x, y ), let T1 be a tangent line to that surface that
lies in the plane y y0 and let T2 be a tangent line to that surface that lies in the plane x x0
at the point P( x0 , y0 , z0 ).
tangent plane, N is
i j k
N v2 v1 0 1 Fy Fx ( x0 , y0 , z0 ) i Fy ( x0 , y0 , z0 ) j k .
1 0 Fx
If R ( x, y, z ) is any point on the tangent plane, then the vector equation of the tangent plane is
PR N 0 and the normal line to the surface through P has the same direction as N.
16
Definition 5.3.2
(i) The tangent plane to S at P is the plane passing through P with normal vector
N F ( P).
(ii) The normal line to S at P is the line passing through P having the same direction as
N F ( P).
Example 5.3.2
1. Find the equation of the tangent line and normal line to the elliptic paraboloid
z 2 x 2 y 2 at the point (1,1,3).
2. Find the points where the normal line to the surface in question 1 meets the xy
plane, xz plane and the yz plane.
Solutions:
1. F ( x, y, z) z 2x2 y 2 Fx ( x, y, z) 4 x Fx (1,1,3) 4
N 4i 2 j k
4( x 1) 2( y 1) 1( z 3) 0 4 x 2 y z 3
x 1 y 1 z 3
4 2 1
2. In the xy plane, z 0
z 3 t 0 3 t t 3
x 1 4t 1 4(3) 13
y 1 2t 1 2(3) 7
Thus, the normal line to the elliptic paraboloid meets the xy plane at (13, 7, 0).
17
In the xz plane, y 0
1
y 1 2t 0 1 2t t
2
1
x 1 4t 1 4 1
2
1 7
z 3 t 3
2 2
Thus, the normal line to the elliptic paraboloid meets the xz plane at (1, 0, 72 ).
In the yz plane, x 0
1
x 1 4t 0 1 4t t
4
1 1
y 1 2t 1 2
4 2
1 13
z 3t 3
4 2
Thus, the normal line to the elliptic paraboloid meets the yz plane at (0, 12 , 132 ).
Suppose we wish to differentiate a function of several variable whose variables are expressed
in terms of other independent variables. In this case, we use chain rules to differentiate such a
function.
Theorem 5.4.1
dx
Let z f ( x, y ) be differentiable and suppose that x x (t ) and y y (t ). Assume that and
dt
dy
exist and are continuous. Then, z can be written as a function of the parameter t , and
dt
dz z dx z dy
dt x dt y dt
18
Theorem 5.4.2
z z x z y
r x r y r
and
z z x z y
s x s y s
Example 5.4.1
dz
1. Let z xy 2 , where x(t ) cos t and y (t ) sin t. Calculate .
dt
w w
2. Suppose that w x3 y 2 , where x(r , s) rs 2 and y (r , s) s 2 sin r. Find and .
r s
Solutions:
z z
1. z xy 2 y 2 and 2 xy
x y
dx dy
x(t ) cos t
sin t and y (t ) sin t cos t.
dt dt
dz z dx z dy
y 2 sin t 2 xy cos t sin 2 t.sin t 2 cos t.sin t.cos t
dt x dt y dt
sin 3 t 2 cos 2t sin t
w w
2. w x3 y 2 3x 2 and 2y
x y
x x y y
x(r , s ) rs 2 s2 , 2rs and y (r , s ) s 2 sin r s 2 cos r , 2s sin r.
r s r s
w w x w y
3x 2 ( s 2 ) 2 y ( s 2 sin r )
r x r y r
3(rs 2 ) 2 ( s 2 ) 2( s 2 sin r )( s 2 sin r )
3r 2 s 6 2s 4 sin 2 r
19
w w x w y
3x 2 (2rs) 2 y (2 s sin r )
s x s y s
3(rs 2 ) 2 (2rs) 2( s 2 sin r )(2 s sin r )
6r 3 s 5 4s 3 sin 2 r.
The chain rules given in Theorem 5.4.1 and Theorem 5.4.2 can be extended to functions of
more than two variables.
Theorem 5.4.3
dx dy
Let w f ( x, y, z ) be a differentiable function. If x x (t ) , y y (t ) , z z (t ) and if ,
dt dt
dz
and exist and are continuous, then
dt
dw w dx w dy w dz
.
dt x dt y dt z dt
x x y y z z
Similarly, if x x(r , s ) , y y (r , s ) , z z (r , s ) and if , , , , and all exist
r s r s r s
and are continuous, then,
w w x w y w z
r x r y r z r
and
w w x w y w z
.
s x s y s z s
w w x w y w z
,
r x r y r z r
w w x w y w z
s x s y s z s
and
w w x w y w z
.
t x t y t z t
20
Example 5.4.2
dw
1. Let w xy z 2 , where x et , y t and z cosh t. Find .
dt
w w
2. Let w xy z 2 , where x 1 ln( s t ), y r s t and z 2r s t. Find ,
r s
w
and
t
Solutions:
w w w
1. w xy z 2 y, x and 2 z.
x y z
dx dy dz
x et et , y t 1 and z cosh t sinh t.
dt dt dt
dw w dx w dy w dz
y (et ) x(1) 2 z (sinh t )
dt x dt y dt z dt
tet et 2 cosh t sinh t
x x 1 x
2. x 1 ln( s t ) 0,
r s s t t
y y y
y r s t 1
r s t
z z z
z 2r s t 2r s ln 2 and 1.
r s t
w w x w y w z
0 x(1) 2 z (2r s ln 2)
r x r y r z r
1 ln( s t ) 2 2r s t (2r s ln 2)
1 ln( s t ) 22 r 2 s 1 ln 2 2r s t ln 2.
w w x w y w z 1 r s
y x(1) 2 z (2 ln 2)
s x s y s z s st
1
1 ln( s t ) 2 2 t (2 ln 2)
r s r s
(r s t )
st
r st
1 ln( s t ) 2 2 r 2 s 1 ln 2 2 r s 1 t ln 2.
st
and
w w x w y w z 1
y x(1) 2 z (1).
t x t y t z t st
1 rs
(r s t ) 1 ln( s t ) 2(2 t )
st
r st
1 ln( s t ) 2r s 1 2t
st
21
Theorem 5.4.3
dy F ( x, y )
x .
dx Fy ( x, y )
Theorem 5.4.3
z F ( x, y , z ) z Fy ( x, y, z )
x and .
x Fz ( x, y, z ) y Fz ( x, y, z )
Example 5.4.3
Solutions:
dy F ( x, y) 12 x 2 5 12 x 2 5
1. x
dx Fy ( x, y) 4 y3 3 4 y3 3
z F ( x, y , z ) 2 xz 2 y 2
2. x 2
x Fz ( x, y, z ) 2 x z 3z 2 4 y
and
z Fy ( x, y, z ) 2 xy 4 z
2
y Fz ( x, y, z ) 2 x z 3z 2 4 y
22
5.5. TOTAL DIFFERENTIALS AND APPROXIMATION
dy f ( x)dx.
Now, for a functions of two variables z f ( x, y ), with x and y denoting the increments
of x and y respectively, the increment of z is given by
z f ( x x, y y ) f ( x, y ).
Definition 5.5.1
dx x and dy y
z z
dz dx dy f x ( x, y )dx f y ( x, y )dy.
x y
Remark 5.5.1
Definition 5.5.1 can be extended to a function of more than two variables. For example, if
w f ( x, y, z ), then
dx x, dy y and dz x
w w w
dw dx dy dz f x ( x, y, z )dx f y ( x, y, z )dy f z ( x, y, z )dz.
x y z
Example 5.5.1
2.98 4.03
2 2
2. Use total differentials to estimate .
23
3. The radius of a cone is 15 cm and the height is 25 cm. There is a maximum error of
0.02 cm in the measurement of the radius and 0.05cm in the measurement of the height.
Solutions:
(b) If w x 2 y 2 z 2 , then
z f ( x, y) x2 y 2
x x
dz f x ( x, y)dx f y ( x, y)dy dx dy
x2 y 2 x2 y 2
At ( x, y ) (3, 4),
3 3
dz (0.02) (0.03) 0.012
3 4
2 2
3 42
2
1 1
V r 2 h (15) 2 (25) 1875 .
3 3
24
(b) The maximum error is V dV , with r dr 0.02 and h dh 0.05
V dV Vr (r , h)dr Vh (r , h)dh
2 1
rh dr r 2 dh
3 3
2 1
(15)(25) (0.02) (15)2 (0.05)
3 3
27.5 cm3
We start with the definition of extrema of functions of two variables. For a function of more
than two variables, the same definition can be extended.
Definition 5.6.1
f ( x, y ) f ( x0 , y0 ).
f ( x, y ) f ( x0 , y0 ).
extreme point of f .
Definition 5.6.2
f ( x0 , y0 ) 0.
25
Remark 5.6.1
Definition 5.6.2 implies that to find critical point(s) of a function of two variables, we need to
solve simultaneously the pair
f f
0 and 0.
x y
The fact that a point ( x0 , y0 ) is a critical point does not guarantee that it is an extreme point of
f.
Definition 5.6.3
A critical point of a function f that is not an extreme point of f is called a saddle point.
Remark 5.6.2
An absolute (or global) maximum or minimum of a function may occur at a boundary point of
its domain.
Example 5.6.1
1. For a function
f ( x, y ) 1 x 2 3 y 2 ,
f f
0 and 0 implies that 2x 0 and 6 y 0 showing that (0, 0) is the only
x y
critical point of f . At this critical point, f has a local (and global) minimum which is
f (0, 0) 1.
f f
2. If f ( x, y ) x 2 y 2 , then 0 2 x 0 x 0 and 0 2y 0 y 0
x y
(0, 0) is a critical point.
But f (0, 0) 0 is neither a local minimum nor a local maximum since
if | x | | y |, f ( x, y ) 0 and | x | | y |, f ( x, y ) 0.
Test for Extrema: Let f and all its first and second partial derivatives be continuous in a
neighbourhood of the critical point ( x0 , y0 ). Let
D( x0 , y0 ) f xx ( x0 , y0 ) f yy ( x0 , y0 ) f xy ( x0 , y0 ) .
2
26
(i) If D 0 and f xx ( x0 , y0 ) 0, then f has a local minimum at ( x0 , y0 ).
Example 5.6.2
Solutions:
1. f ( x, y) x 2 4 xy y 3 4 y f x ( x, y) 2 x 4 y and f y ( x, y) 4 x 3 y 2 4
2 x 4 y 0 and 4 x 3 y 2 4 0
4 2
Solving these simultaneously, we obtain the pairs (4, 2) and , .
3 3
f xx ( x, y) 2, f yy ( x, y) 6 y and f xy ( x, y) f yx ( x, y) 4.
Thus,
4 2
Thus, , is a saddle point.
3 3
2. f ( x, y) 4 x 2 4 xy y 2 5 f x ( x, y) 8x 4 y and f y ( x, y) 4 x 2 y
27
D( x, y ) f xx ( x, y ) f yy ( x, y ) f xy ( x, y ) 8(2) ( 4) 2 0.
2
Thus, by the test for extrema, every point on y 2 x can be a local minimum, local maximum
or a saddle point. Notice, however, that
f ( x, y ) 4 x 2 2 xy 2 xy y 2 y 2 5
2 x(2 x y ) y (2 x y ) 5
(2 x y ) 2 5
THE END!
28