Mathematics II

Week 4
 This week’s topic
Functions of two variables
 Functions of multiple variables
A function describes a relationship between the so-called independent variable and the so-called dependent
variable. However, the independent variable sometimes is more than just one variable. For example, the area of
a rectangle depends on two variables: its length (or rather, its, height) and its width. The surface of a land, for
example, can be described as a function that takes the height as a value at each point (given by two coordinates –
thus having two independent variables) of the “plane” laid at sea level.

A function f with two variables (x and y) is denoted by f (x, y) (where f maps the plane into the set of real numbers,
f : R2 → R) we can picture in the space (using GeoGebra, WolframAlpha, Maple, etc.).

Remark. A function can have more than just one or two independent variables.
 Motivation
Economic processes usually cannot be described with functions having only one variable. So we are going to
study two variable functions. (Multivariable fuctions can be studied based on these.) Our aims are as follows.
Knowing the value of a function at a point (a, b), we want to be able to tell
1. the approximate values of the function for points “close to” (a, b);
2. what effect the change of one of the variables has on the other variable if we allow no change (or only slight
change) in the value of the function;
3. how we can get the most increase in the function’s value with little change.

For 1. we will give a “linear” approximation of the function at a given point, which means that we will lay a plane
tangent to the graph of the function at the given point.
For 2. we will have to study the points (x, y) for which f (x, y) is equal.
For 3. we will have to find in which direction the change of the graph of the function is the steepest.
 Slicing the surface – the contour-lines . . .
We can cut through the surface with a plane parallel to the x-y plane at
a certain value (say, c) given as z = c.

The intersection of the cutting plane and the surface is called a contour
line. It’s a set of points in the plane: {(x, y, c) | f (x, y) = c}. Sometimes
we project the contour lines to the x-y plane to make it easier to handle
them. {(x, y) | f (x, y) = c}.

Contour lines are used e.g. in topography.

Contour lines are not necessarily lines in geometric sense.
 . . . and cross sections of two variable functions’ graphs
If we fix y at b, the plane y = b cuts through the surface creating a function of one variable only: x ↦→ f (x, b).
Similarly, the plane x = a cutting through the surface creates a function depending on y only: y ↦→ f (a, y).

Example. f (x, y) = 2xy + y. For y = 1 we have x ↦→ f (x, 1) = 2x + 1; and for x = 0.5 we have y ↦→ f (0.5, y) = y + y = 2y.
 Partial derivatives1
We would like to give a linear approximation to the function f – that is a
tangent plane of the graph – at point (a, b, f (a, b)).
The tangent plane of a function at a given point is tangent in all directions.
Therefore, it is enough to find two tangent lines and get the plane spanned
by them.
Let f (x, y) be a two variable function. Let the point (a, b) be in the x-y
plane (in the domain of the function f ).
x ↦→ f (x, b) and y ↦→ (a, y) are one variable functions. We know how to
take the derivative of such functions.
The so-called partial derivative of f with respect to x is the derivative of
f when y is fixed (held constant, y = b), which is denoted by fx′ .
The partial derivative of f with respect to y is the derivative of f when x
is fixed (held constant, x = a), which is denoted by fy′ .
𝜕f d f 𝜕f d f
Remark. fx′ has other notations, like 𝜕x f , , . Similarly, fy′ can also be denoted by 𝜕y f , , .
𝜕x d x 𝜕y d y

1For simplicity, we are only going to deal with functions that have derivatives.
 Example for partial derivatives
Consider the two variable function f (x, y) = x2 + xy + y2 .
Fixing y , the derivative of f (x, y) = x2 + x y + y2 with respect to x is 2x + y .
Fixing x , the derivative of f (x, y) = x2 + x y + y2 with respect to y is x + 2y.

For example, let x = 3, y = −0.5.

The slopes of the tangent lines in the cross section planes at point (3, −0.5) are
fx′ (3, −0.5) = 6.5 and fy′ (3, −0.5) = 4.
 The two tangent lines of x2 + xy + y2 at (3, −0.5)

The two tangent lines of the two cross sections determine a plane. fx′ and fy′ are actually two-variable functions.
They take a value at (a, b). These values are the slopes of the two tangent lines. The slope in one direction (parallel
to the x-z plane) for point (a, b) is fx′ (a, b), and the slope in the other direction (parallel to the y-z plane) is fy′ (a, b).
 The equation of the tangent plane of f at (a, b, f (a, b))
The point (a, b, f (a, b)) is a point on the surface of f (in the space).
The tangent plane of f at this point has slope fx′ (a, b) in one direction (parallel to the x-z plane) and slope fy′ (a, b)
in another direction (parallel to the y-z plane).

The equation of the tangent plane at point (a, b, f (a, b)) has the form
z − f (a, b) = fx′ (a, b) · (x − a) + fy′ (a, b) · (y − b)
or
z = fx′ (a, b) · (x − a) + fy′ (a, b) · (y − b) + f (a, b)
or
0 = fx′ (a, b) · (x − a) + fy′ (a, b) · (y − b) − [z − f (a, b)].
 Example
Let f (x, y) = x2 − 2x2 y2 + y2 , and let (a, b) = (1, −2).
Find the equation of the tangent plane of the function f at (a, b).
Solution. We use that
z = f (a, b) + fx′ (a, b) · (x − a) + fy′ (a, b) · (y − b)

f (1, −2) = 1 − 8 + 4 = −3.

fx′ = 2x − 4xy2 ; fx′ (1, −2) = 2 · 1 − 4 · 1 · 4 = −14
fy′ = −4x2 y + 2y; fy′ (1, −2) = −4 · 1 · (−2) + 2 · (−2) = 4
z = −3 + (−14) · (x − 1) + 4 · (y + 2) − 3 = −14x + 4y + 19
 Linear approximation of the function f (x, y)
We found that the tangent plane of f at point (a, b, f (a, b)) is
z − f (a, b) = fx′ (a, b) (x − a) + fy′ (a, b) (y − b).
This will help us to give an approximation of the values of f for points close to (a, b). (Just as in the case of one
variable functions.)
If (x, y) is close to (a, b), then f (x, y) is close to the point corresponding to it on the tangent plane:
f (x, y) − f (a, b) ≈ fx′ (a, b) (x − a) + fy′ (a, b) (y − b),
which can be written in short notation as
Δf ≈ fx′ · Δx + fy′ · Δy.

For example, for f (x, y) = x2 − 2x2 y2 + y2 at point (a, b) = (1, −2) we get the approximation f (x, y) ≈ −14x + 4y + 19.
Therefore, f (1.1, −1.9) ≈ −15.4 − 7.6 + 19 = −4, while, in fact, f (1.1, −1.9) = −3.9162.

With this we have given the answer to the first problem posed in “Motivation”.
 Example
A company produces sweat suits (pants and shirts). They are sold separately but, of course, some buy them in a
set.
The unit price of the pants is x = $15, and that of the shirts is y = $12 apiece.
The daily demand of these in terms of the prices is D1 (x, y) = 12 + 3xy − 4x for the pants and D2 (x, y) = 3xy − 3y
for the shirts.
(a) What is the total revenue after the two products?
(b) Use linear approximation to estimate the change of the revenue if they increase the unit price of the pants by
$1 and decrease the price of the shirts by $1?
(a) R(x, y) = R1 (x, y) + R2 (x, y) = 12x + 3x2 y − 4x2 + 3xy2 − 3y2 .
R(15, 12) = 12 · 15 + 3 · 225 · 12 − 4 · 225 + 3 · 15 · 144 − 3 · 144 = 13 428.
(b) Linear approximation: ΔR ≈ Rx′ (a, b) · Δx + Ry′ (a, b) · Δy = Rx′ (15, 12) · (1) + Ry′ (15, 12) · (−1).
Rx′ (x, y) = 12 + 6xy − 8x + 3y2 ; Rx′ (15, 12) = 12 + 6 · 15 · 12 − 8 · 15 + 3 · 144 = 1404.
Ry′ (x, y) = 3x2 + 6xy − 6y; Ry′ (15, 12) = 3 · 225 + 6 · 15 · 12 − 6 · 12 = 1683.
ΔR ≈ 1404 · (1) + 1683 · (−1) = − 279
 Approximating values of a function on a contour line
We would like to know how much y should be changed from b when x changes from a provided the function value
remains f (a, b), that is, we stay on the same contour line.

In the approximation Δf ≈ fx′ · Δx + fy′ Δy, Δf must be 0.

fx′ (a, b)
0 ≈ fx′ (a, b)(x − a) + fy′ (a, b) (y − b) means that y ≈ − · (x − a) + b.
fy′ (a, b)
Using other notations: when there is no change of the function (we are on the same contour line),
fx′ (a, b)
Δy ≈ − Δx.
fy′ (a, b)
Of course, we can estimate the change in x in terms of the change of y by rearranging it:
fy′ (a, b)
Δx ≈ − Δy.
fx′ (a, b)

For this estimate, we used the tangent plane of f at (a, b).

We could, however, give a similar approximation with the help of the tangent line of the contour line.
 Tangent line of a contour line of f
Suppose we have a line l in the tangent plane at points (a, b) that also goes through this point. l is a tangent line
itself to the surface. One of these tangent lines is parallel to the x-y plane. Since it lies in the plane of the contour
line, it is also in the plane z = f (a, b).
Now we have two planes (the tangent plane and the plane of the contour line) that both contain the tangent line in
question. Thus, for the intersection of the two planes we get
f (a, b) = f (a, b) + fx′ (a, b) (x − a) + fy′ (a, b) (y − b)
fx′ (a, b) fx′ (a, b)
Subtracting f (a, b) and dividing by fy′ (a, b), we get 0 = · (x − a) + y − b. Let’s denote − by m, so
fy′ (a, b) fy′ (a, b)
y = m(x − a) + b or y − b = m(x − a).

fx′ (a, b)
The quotient m = − is called the implicite derivative of f at (a, b).
fy′ (a, b)
We call it implicite because the contour line is not necessarily a graph of a function and doesn’t have to have a
derivative, but it can have a tangent given implicitly.
The contour line is in the space. We project it onto the x-y plane. The expression y = m(x − a) + b is actually the
plane along the projection goes on.
 An image of the tangent line of the contour line

The tangent plane (yellow) and the contour line’s plane (gray) intersect in a tangent line to the contour line.
y = m(x − a) + b is actually a plane (green). It is perpendicular to the x-y plane. The intersection of the two planes
is the projected image of the tangent line in question.
fx′ (a, b)
The slope of this line (in the x-y plane) at the point of tangency is − .
fy′ (a, b)
 We learnt. . .

Tangent plane at (a, b, f (a, b)): p(x, y) = f (a, b) + fx′ (a, b) (x − a) + fy′ (a, b) (y − b)
Approximation close to (a, b): f (x, y) ≈ f (a, b) + fx′ (a, b) (x − a) + fy′ (a, b) (y − b); Δf ≈ fx′ Δx + fy′ Δy
fx′ (a, b)
Tangent line at (a, b) on the contour line: y = − (x − a) + b; approximation close to (a, b): Δy ≈ mΔx
fy′ (a, b)
 Example
Let f (x, y) = x2 − 2x2 y2 + y2 , and let (a, b) = (1, −2).
Check out the function value at (a, b), and find the tangent to the contour line at point (a, b, f (a, b)).
Solution. f (1, −2) = 1 − 2 · 1 · 4 + 4 = −3. We use the formula
fx′ (a, b)
y=− (x − a) + b.
fy′ (a, b)
fx′ = 2x − 4xy2 ; fx′ (1, −2) = 2 · 1 − 4 · 1 · 4 = −14
fy′ = −4x2 y + 2y; fy′ (1, −2) = −4 · 1 · (−2) + 2 · (−2) = 4
−14
y=− (x − 1) + 2 = 3.5x − 1.5
4
 A practical example
A company produces sweat suits (pants and shirts). They are sold separately but, of course, some buy them in a
set. The unit price of the pants is x = $15, and that of the shirts is y = $12 apiece.
The daily demand of these in terms of the prices is D1 (x, y) = 12 + 3xy − 4x for the pants and D2 (x, y) = 3xy − 3y
for the shirts.
By how much should we change the unit price of the shirts in order to maintain the revenue?
Rx′ (a, b)
We use linear approximation: Δy ≈ − Δx. Here, Δx = 1, and we need to find Δy.
Ry′ (a, b)
R(x, y) = R1 (x, y) + R2 (x, y) = xD1 (x, y) + yD2 (x, y) = 12x + 3x2 y − 4x2 + 3xy2 − 3y2 .
Rx′ (x, y) = 12 + 6xy − 8x + 3y2 ; Rx′ (15, 12) = 12 + 6 · 15 · 12 − 8 · 15 + 3 · 144 = 1404.
Ry′ (x, y) = 3x2 + 6xy − 6y; Ry′ (15, 12) = 3 · 225 + 6 · 15 · 12 − 6 · 12 = 1683.
Rx′ (15, 12)
Rx′ (15, 12) = 1404 and Ry′ (15, 12) = 1683, Δy ≈ − ≈ −0.8342246.
Ry′ (15, 12)
We should decrease the unit price of the shirts by (at most about) 83 cents.
 The steepest slope of the surface
If, for example, the profit was given as a two variable function, we might want to know how we can increase our
profit.
We want to find out that starting from point (a, b, f (a, b)) which direction do we have to go to get the highest
increase on the surface of f .
Moving away from (a, b) by a unit vector (u, v) (for which u2 + v2 = 1), the difference in the values of f is:
f (a + u, b + v) − f (a, b) ≈ fx′ (a, b) · (a + u − a) + fy′ (a, b) · (b + v − b) =
= fx′ (a, b) · (u) + fy′ (a, b) · (v).

It is the dot product of the so-called gradient vector (fx′ (a, b), fy′ (a, b)) and a unit vector (u, v). The length of
these two vectors do not change. The dot product of two such vectors depend on the angle enclosed by the two
vectors (it is the cosine of it) and takes its maximum when the angle is 0. Thus, (u, v) has the same direction as
(fx′ (a, b), fy′ (a, b)).

The greatest increase in f can be achieved in the direction of the vector (fx′ (a, b), fy′ (a, b)).
This answers our third and last question.
 Example
Let f (x, y) = x2 − 2x2 y2 + y2 , and let (a, b) = (1, −2).
Find the steepest slope of f at (a, b, f (a, b)).
Solution. fx′ = 2x − 4xy2 ; fx′ (1, −2) = 2 · 1 − 4 · 1 · 4 = −14
fy′ = −4x2 y + 2y; fy′ (1, −2) = −4 · 1 · (−2) + 2 · (−2) = 4
The steepest slope has direction (−14, 4).
 This week we learnt. . .
Two variable (multivariable) function: the domain is in R2 , the range is in R.
Cross sections of the graph of a two variable functions;
contour line (cross section with a plane parallel to the x-y plane);
partial derivatives (derivative with respect to a particular variable);
tangent plane of a surface;
linear approximation of a two variable function on the tangent plane (with the help of partial derivatives);
tangent line of the contour line;
implicite derivative: the slope of the tangent of the contour line in the x-y plane;
linear approximation on a tangent line of a contour line;
the steepest slope’s direction on the surface of a function at a given point.
 How to. . . A guide to the exercises
Identify the function (in case of a word problem) and the point given: f (x, y) and (a, b).
What to calculate, what to use:
fx′ (a, b)
f (a, b), fx′ (x, y), fy′ (x, y), fx′ (a, b), fy′ (a, b), m = −
fy′ (a, b)
Approximating the function at (x, y) close to (a, b):
f (a, b) ≈ f (x, y) + fx′ (a, b) · (a − x) + fy′ (a, b) · (b − y)
f (a, b) − f (x, y) ≈ fx′ (a, b) · (a − x) + fy′ (a, b) · (b − y)
or Δf ≈ fx′ Δx + fy′ Δx

On the contour line

fx′ (a, b)
y−b = − (x − a)
fy′ (a, b)
f′
or Δy ≈ − x′ Δx.
fy

The tangent of f (the direction of the steepest slope) at point (a, b) is (fx′ (a, b), fy′ (a, b)).
 Example 4.1.1
y
Let f (x, y) = . Determine the equation of the contour line of f at (3; −6).
x
Find the partial derivatives, give a linear approximation at point (3; −6)

−6 y
(a, b) = (3, −6), f (3, −6) = = −2 . The contour line: = −2, that is y = −2x.
3 x
y −6 2 1 1
fx′ (x, y ) = − 2 , fx′ (3, −6) = − 2 = ; fy′ ( x , y) = , fy′ (3, −6) = .
x 3 3 x 3
The tangent plane:
2 1 2 1
z = fx′ (x, y)(x − a) + fy′ (x, y) (y − b) + f (a, b) = (x − 3) + (y + 6) − 2 = x + y − 2
3 3 3 3
The linear approximation of f close to (a, b):
2 1
f (x, y) ≈ fx′ (a, b) (x − a) + fy′ (a, b) (y − b) + f (a, b) = x + y − 2
3 3
2
fx′ (a, b) 3
Implicite derivative: m = − = − = −2.
fy′ (a, b) 1
3
Tangent line of contour line at (3, −6): y = m(x − a) + b = −2(x − 3) − 6 = −2x.
 Example 4.1.2
Determine the contour line of f (x, y) = x2 y at point (a, b) = (1, 3); find the partial derivatives; give a linear
approximation of f at point (1, 3), find the equation of the tangent line of f at (1; 3) (on the contour line).

3
f (1, 3) = 12 · 3 = 3. Contour line: x2 y = 3, y= .
x2
fx′ (x, y) = 2xy, fx′ (1, 3) = 2 · 1 · 3 = 6; fy′ (x, y) = x2 , fy′ (1, 3) = 1.
Linear approximation of f : f (x, y) ≈ 6(x − 1) + (y − 3) + 3 = 6x + y − 6.

6
Implicite derivative: − = −6.
1
The tangent line of the contour line at (1; 3): y = −6(x − 1) + 3 = −6x + 9.
 Example 4.1.3 – linear approximation
Production: Q(x, y) = 0.08x2 + 0.12xy + 0.03y2 pieces/day; x: skilled workers’ hours, y: unskilled workers’ hours.
(x, y) = (80, 200). x increases by 1, 2, resp.
(a) What is the change of Q? (b) How to change y to maintain the level of production for x = 1?

(a) Linear approximation for the change: ΔQ = Qx′ (a, b)Δx + Qy′ (a, b)Δy. Δx = 1, 2, resp., Δy = 0.
Qx′ (x, y) = 0.16x + 0.12y, Qx′ (a, b) = Qx′ (80, 200) = 12.8 + 24 = 36.8;
Qy′ (x, y) = 0.12x + 0.06y, Qy′ (a, b) = Qy′ (80, 200) = 9.6 + 12 = 21.6.
ΔQ = 36.8 · 1 + 21.6 · 0 = 36.8 and ΔQ = 36.8 · 2 + 21.6 · 0 = 73.6.

Qx′ (a, b)
(b) Δy = − Δx, where Qx′ (a, b) = 36.8, Qy′ (a, b) = 21.6;
Qy′ (a, b)
36.8
For Δx = 1, Δy = − ≈ −1.7. For Δx = 2, Δy = −3.4.
21.6
 Exercises/4.2.1
Please, practice the theoretical background by solving Exercise 4.2.1.
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
6. Check the contour plot of the graph of the function in WolframAlpha. (Or use GeoGebra.)
 Exercises/4.2.1.
(a) f (x, y) = x + y2 , (a, b) = (1, 2) (i) f (x, y) = xy, (a, b) = (−3, 2)
(b) f (x, y) = x2 + y2 , (a, b) = (3, −1) (j) f (x, y) = ye−xy , (a, b) = (1, 2)

(c) f (x, y) = 3x − 2y + 5, (a, b) = (−3, 2) (k) f (x, y) = xy2 , (a, b) = (3, −1)
√︃
(d) f (x, y) = 5x − 3y, (a, b) = (−1, 2) (l) f (x, y) = x2 + y2 , (a, b) = (3, −4)
(e) f (x, y) = x2 + x − 4y, (a, b) = (3, −1) x
(m) f (x, y) = , (a, b) = (2, −1)
y
(f) f (x, y) = x(y − x3 ), (a, b) = (1, 1)
(n) f (x, y) = yx , (a, b) = (3, 2)
(g) f (x, y) = xy − y(y + 1), (a, b) = (1, 0)
(o) f (x, y) = ln(−x + 3y2 ), (a, b) = (−3, 2)
s √
(h) f (r, s) = , (a, b) = (e2 , 3) (p) f (x, y) = x2 y3 , (a, b) = ( 2, 2)
ln r
 Solution 4.2.1. (a) f (x, y) = x + y2 , (a, b) = (1, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (1, −2) = 1 + 22 = 5
2. fx′ (x, y) = 1, fy′ (x, y) = 2y
3. fx′ (1, 2) = 1, fy′ (1, 2) = 4
fx′ (x, y) 1
4. m = − ′ =−
fy (x, y) 4
1 1 9
5. y = m(x − a) + b = − (x − 1) + 2 = − x +
4 4 4
 Solution 4.2.1. (b) f (x, y) = x2 + y2 , (a, b) = (3, −1)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (3, −1) = 32 + (−1) 2 = 10
2. fx′ (x, y) = 2x; fy′ (x, y) = 2y
3. fx′ (3, −1) = 6; fy′ (3, −1) = − 2
fx′ (3, −1)
4. m = − =3
fy′ (3, −1)
5. y = 3(x − 3) + (−1) = 3x − 10
 Solution 4.2.1. (c) f (x, y) = 3x − 2y + 5, (a, b) = (−3, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (−3, 2) = − 9 − 4 + 5 = − 13
2. fx′ (x, y) = 3; fy′ (x, y) = − 2
3. fx′ (−3, 2) = 3; fy′ (−3, 2) = − 2
fx′ (−3, 2) 3
4. m = − =−
fy′ (−3, 2) 2
3 3 13
5. y = (x + 3) + 2 = x +
2 2 2
 Solution 4.2.1. (d) f (x, y) = 5x − 3y, (a, b) = (−1, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (−1, 2) = − 5 − 6 = − 11
2. fx′ (x, y) = 5; fy′ (x, y) = − 3
3. fx′ (−1, 2) = 5; fy′ (−1, 2) = − 3
fx′ (−1, 2) 5
4. m = − =
fy′ (−1, 2) 3
5 5 11
5. y = (x + 1) + 2 = x +
3 3 3
 Solution 4.2.1. (e) f (x, y) = x2 + x − 4y, (a, b) = (3, −1)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (3, −1) = 9 + 3 + 4 = 16
2. fx′ (x, y) = 2x + 1; fy′ (x, y) = − 4
3. fx′ (3, −1) = 7; fy′ (3, −1) = − 4
fx′ (3, −1) 7
4. m = − =
fy′ (3, −1) 4
7 7 25
5. y = (x − 3) + (−1) = x −
4 4 4
 Solution 4.2.1. (f) f (x, y) = x(y − x3 ), (a, b) = (1, 1)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (1, 1) = 1 · (1 − 1) = 0
2. fx′ (x, y) = y − 4x3 ; fy′ (x, y) = x
3. fx′ (1, 1) = − 3; fy′ (1, 1) = 1
fx′ (1, 1)
4. m = − =3
fy′ (1, 1)
5. y = 3(x − 1) + 1 = 3x − 2
 Solution 4.2.1. (g) f (x, y) = xy − y(y + 1), (a, b) = (1, 0)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (1, 0) = 1 · 0 − 0 · 1 = 0
2. fx′ (x, y) = y; fy′ (x, y) = x − 2y − 1
3. fx′ (1, 0) = 0; fy′ (1, 0) = 0
fx′ (1, 0)
4. m = − = cannot be told
fy′ (1, 0)
This doesn’t give us a slope. We have to take one step backwards.
What is the equation of the contour line?
f (x, y) = 0, that is, xy = y(y + 1).
Either y = 0 or x = y + 1, that is, y = x − 1.
This is a set of two lines (not a function). It has two tangent lines, the lines themselves.
 s
Solution 4.2.1. (h) f (r, s) = , (a, b) = (e2 , 3)
ln r
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
3 3
1. f (e2 , 3) = =
ln e2 2
s 1 1
2. fr′ (r, s) = − 2
· ; fs′ (r, s) =
(ln r) r ln r
3 1
3. fr′ (e2 , 3) = − 2 2 ; fs′ (e2 , 3) =
2 ·e 2
fr′ (e2 , 3) 3
4. m = − ′ 2 =
fs (e , 3) 2 · e2
3 3 3
5. s = 2 (r − e2 ) + 3 = 2 r +
2e 2e 2
 Solution 4.2.1. (i) f (x, y) = xy, (a, b) = (−3, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (−3, 2) = − 6
2. fx′ (x, y) = y; fy′ (x, y) = x
3. fx′ (−3, 2) = 2; fy′ (−3, 2) = − 3
fx′ (−3, 2) 2
4. m = − =
fy′ (−3, 2) −3
2 2
5. y = (x − (−3)) + 2 = x + 4
3 3
 *Solution 4.2.1. (j) f (x, y) = ye−xy , (a, b) = (1, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
2
1. f (1, −2) = 2 · e−2 = 2
e
2. fx′ (x, y) = − y2 · e−xy ; fy′ (x, y) = e−xy + ye−xy · (−x) = (1 − xy)e−xy
4 1
3. fx′ (1, 2) = − 4 · e−2 = − ; fy′ (1, 2) = − e−2 = − 2
e2 e
fx′ (1, 2)
4. m = − = −4
fy′ (1, 2)
5. y = − 4(x − 1) + 2 = − 4x + 6
 Solution 4.2.1. (k) f (x, y) = xy2 , (a, b) = (3, −1)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (3, −1) = 3 · (−1) 2 = 3
2. fx′ (x, y) = y2 ; fy′ (x, y) = 2xy
3. fx′ (3, −1) = 1; fy′ (3, −1) = −6
fx′ (3, −1) 1
4. m = − =
fy′ (3, −1) 6
1 1 3
5. y = (x − 3) − 1 = x −
6 6 2
 √︃
Solution 4.2.1. (l) f (x, y) = x2 + y2 , (a, b) = (3, −4)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
√
1. f (3, −4) = 9 + 16 = 5
1 1 x 1 1 y
2. fx′ (x, y) = · √︁ · 2x = √︁ ; fy′ (x, y) = · √︁ · 2y = √︁
2 x2 + y2 2
x +y 2 2 2
x +y 2 x + y2
2

3 4
3. fx′ (3, −4) = ; fy′ (3, −4) = −
5 5
fx′ (3, −4) 3
4. m = − ′ =
fy (3, −4) 4
3 3 25
5. y = (x − 3) − 4 = x −
4 4 4
 x
Solution 4.2.1. (m) f (x, y) = , (a, b) = (2, −1)
y
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (2, −1) = − 2
1 x
2. fx′ (x, y) = ; fy′ (x, y) = − 2
y y
3. fx′ (2, −1) = − 1; fy′ (2, −1) = −2
fx′ (2, −1) 1
4. m = − =−
fy′ (2, −1) 2
1 1
5. y = − (x − 2) − 1 = − x
2 2
 Solution 4.2.1. (n) f (x, y) = yx , (a, b) = (3, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (3, 2) = 9
2. fx′ (x, y) = ln(y) · yx ; fy′ (x, y) = x · yx−1 for x ≠ 0 and 0 for x = 0.
3. fx′ (3, 2) = 8 ln(2); fy′ (3, 2) = 12 or 0
fx′ (3, 2) 2 ln(2)
4. m = − ′ =− (for x ≠ 0)
fy (3, 2) 3
2 ln(2) 2 ln(2)
5. y = − (x − 3) + 2 = − x + 2 ln(2) + 2
3 3
 Solution 4.2.1. (o) f (x, y) = ln(−x + 3y2 ), (a, b) = (−3, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
1. f (−3, 2) = ln(15)
1 1 1 6y
2. fx′ (x, y) = 2
· (−1) = 2
; fy′ (x, y) = 2
· 6y = 2 .
−x + 3y x − 3y −x + 3y 3y − x
1 12
3. fx′ (−3, 2) = − ; fy′ (−3, 2) =
15 15
fx′ (−3, 2) 1
4. m = − ′ =
fy (−3, 2) 12
1 1 9
5. y = (x − (−3)) + 2 = x +
12 12 4
 √
Solution 4.2.1. (p) f (x, y) = x2 y3 , (a, b) = ( 2, 2)
1. Find the value of the given function at the given point.
2. Find the partial derivatives of the given function.
3. Find the values of the partial derivatives at the given point.
4. Give the implicite derivative – the slope of the tangent line of the contour line.
5. Give the equation of the tangent line of the contour line at the given point.
√
1. f ( 2, 2) = 16
2. fx′ (x, y) = 2xy3 ; fy′ (x, y) = 3x2 y2 .
√ √ √
3. fx′ ( 2, 2) = 16 2; fy′ ( 2, 2) = 24
√ √
fx′ ( 2, 2) 2 2
4. m = − √ =−
fy′ ( 2, 2) 3
√ √
2 2 √ 2 2 10
5. y = − (x − 2) + 2 = x+
3 3 3
 Exercises/4.2.2
A paint shop distributes two kinds of special paints. If the first and second paints have unit prices x and y USD per
gallon, respectively, then the corresponding demands are D1 (x, y) = 200 − 10x + 20y and D2 (x, y) = 100 + 5x − 10y
gallons per month.
(a) Determine the monthly revenue in terms of unit prices.
(b) At the moment the paint shop is selling the first and second paints for 21 and 16 USD per gallon, respectively.
What is the monthly revenue?
(c) How will the monthly revenue be approximately changed if we increase the first or the second paint’s price by
1 USD per gallon?
(d) How should the price of the second paint be changed keeping the same revenue if we increase the first paint’s
price by 1 USD?
 Exercises/4.2.2
Paint prices: x, y USD/gallon. Demand of each: D1 (x, y) = 200 − 10x + 20y, D2 (x, y) = 100 + 5x − 10y gallons/month;
x = 21, y = 16. (a) R(x, y) (a month)? (b) R(21, 16)? (c) ΔR if Δx = 1 or Δy = 1? (d) Δx = 1, ΔR = 0, Δy =?.

Monthly revenue:
R(x, y) = xD1 (x, y) + yD2 (x, y) = 200x − 10x2 + 20xy + 100y + 5xy − 10y2 =
= −10x2 − 10y2 + 25xy + 200x + 100y (USD)
x = 21, y = 16 R(21, 16) = −4410 − 2560 + 8400 + 4200 + 1600 = 7230 USD
Δx = 1, Δy = 0 ΔR ≈ Rx′ Δx + Ry′ Δy = Rx′ ; Rx′ (x, y) = −20x + 25y + 200; ΔR ≈ −420 + 400 + 200 = 180.
Δx = 0, Δy = 1 ΔR ≈ Rx′ Δx + Ry′ Δy = Ry′ ; Ry′ = −20y + 25x + 100; ΔR ≈= −320 + 525 + 100 = 305.
Rx′ (21, 16) 108
Δx = 1, ΔR = 0 Δy = − · Δx = − ≈ −0,59; a decrease of 59 cents.
Ry′ (21, 16) 305
 Exercises/4.2.3
A factory’s production is Q(K, L) = 120K 2/3 L1/3 units, where K is the invested capital in terms of 1000 USD and
L is the working hours.
(a) What is the production in case of 125000 USD invested capital and 1000 working hours?
(b) What is going to happen with the production if we reduce the invested capital and working hours by half?
(c) Let us consider invested capital and working hours data from subproblem (a). What is going to happen with
the production if we increase the invested capital by 1000 USD and the working hours to 1005 hours?
(d) Let us consider invested capital and working hours data from subproblem (a). If we increase the invested
capital by 1000 USD how should we change the working hours keeping the same production level?
 Exercises/4.2.3
√︁3
Production: Q(K, L) = 120K 2/3 L1/3 = 120 · (K 2 L) 1/3 = 120 K 2 · L units, K: investment (1000 $), L: working hours.

√︁3
(a) K = 125, 000, L = 1000, Q(K, L) = ? Q(125, 1000) = 120 · 1252 · 1000 = 120 · 250 = 30, 000 units.
√3 √︂
Q(K, L) 120 · K 2 L 3 1 1
(b) K and L reduces to half, how does Q change? = √︁3 = = .
Q(K/2, L/2) 120 · K 2 /4 · L/2 8 2
(c) ΔK = 1, ΔL = 5. ΔQ = ? ΔQ = QK′ · ΔK + Ql′ · ΔL = QK′ · 1 + QL′ · 5;
√︂ √︂
2 −1/3 1/3 1 2/3 −2/3 2
′ 3 L ′ 3 K
QK (K, L) = 120 · · K L = 80 ; QL (K, L) = 120 · · K L = 40 · ;
3 K 3 L2
QK′ (125, 1000) = 160; QL′ (125, 1000) = 10. ΔQ = 160 + 50 = 210
QK′ (125, 1000)
(d) ΔQ = 0, ΔK = 1 (1000 USD), ΔL ≈ ? ΔL ≈ − ΔK
QL′ (125, 1000)
QK′ (125, 1000) 160
ΔL ≈ − ΔK = = 16
QL′ (125, 1000) 10
 Exercises/4.2.4
The production in a factory is Q(x, y) = 0.06x2 + 0.14xy + 0.05y2 pieces per day.
The variables x and y denote the working hours of skilled labourers and unskilled labourers, respectively.
At the moment the skilled and unskilled labourers are working 60 and 300 hours in total per day, respectively.
(a) How should we change the working hours of unskilled labourers keeping the same production level if we
increase the working hours of skilled labourers by 1 hour?
(b) How should we change the working hours of unskilled labourers keeping the same production level if we
decrease the working hours of skilled labourers by 2 hours?
(c) How should we change the working hours of skilled labourers keeping the same production level if we decrease
the working hours of unskilled labourers by 2 hours?
 Exercises/4.2.4
Production: Q(x, y) = 0.06x2 + 0.14xy + 0.05y2 pieces per day.
Skilled workers’ working hour x, unskilled workers’ working hour: y. (x, y) = (60, 300).
(a) ΔQ = 0, Δx = 1, Δy =?;
(b) ΔQ = 0, Δx = −2, Δy =?;
(c) ΔQ = 0, Δy = −2, Δx =?;

Qx′ (x, y) = 0.12x + 0.14y; Qx′ (60, 300) = 49.2; Qy′ (x, y) = 0.14x + 0.1y; Qy′ (60, 300) = 38.4.
49.2 1 38.4
m=− = −1.28125 =− ≈ −0.78
38.4 m 49.2
(a) Δy ≈ mΔx = −1.28125 · 1 ≈ −1.28.
(b) Δy ≈ Δx = −1.28125 · (−2) ≈ 2.5625 ≈ 2.56.
1
(c) Δx ≈ Δy ≈ −0.78 · (−2) ≈ 1.56.
m
 Exercises/4.2.5
In an area the incidence of squirrels depends on the number of oak trees and hazelnut bushes.
Based on a research the number of squirrels is Q(x, y) = 0.2x2 + 0.05xy2 + 0.1y2 , where x and y denote the number
of hazel bushes and oak trees in this area, respectively.
At the moment, there are 20 hazel bushes and 30 oak trees.
If we cut down an oak tree, then how many hazel bushes should be planted keeping the same number of squirrels?
 Exercises/4.2.5
Q(x, y) = 0.2x2 + 0.05xy2 + 0.1y2 , x: hazel; y: oak; (x, y) = (20, 30).
ΔQ = 0, Δy = −1, Δx =?

Qx′ (x, y) = 0.4x + 0.05y2 ; Qx′ (20, 30) = 53;

Qy′ (x, y) = 0.1xy + 0.2y; Qy′ (20, 30) = 66.
Qy′ (20, 30)
Δx ≈ − Δy ≈ 1.245
Qx′ (20, 30)
 Exercises/4.2.6
Two products can be sold at a unit price of p1 and p2 USD.
The corresponding demand functions are D1 (p1 , p2 ) and D2 (p1 , p2 ) (specified demand functions will be given
below).
How will the saleable quantities change if we increase one of the unit prices by 1 USD?
Based on that, think about the relation between the two products. Give an example.
(a) D1 = 500 − 6p1 + 5p2 , D2 = 200 + 2p1 − 5p2
(b) D1 = 1000 − 0.02p21 − 0.05p22 , D2 = 800 − 0.001p21 − p1 p2
 Exercises/4.2.6
Unit prices: p1 , p2 USD. Demand D1 (p1 , p2 ) and D2 (p1 , p2 ). Δp1 = 1 or Δp2 = 1. ΔD1 =?, ΔD2 =?
(a) D1 (p1 , p2 ) = 500 − 6p1 + 5p2 , D2 (p1 , p2 ) = 200 + 2p1 + 5p2 .
(b) D1 (p1 , p2 ) = 1000 − 0.02p21 − 0.05p22 , D2 (p1 , p2 ) = 800 − 0.001p21 − p1 p2 .
What is the relation between the two products?

Case (a) If p1 increases, D1 decreases and D2 increases. If p2 increases, both D1 and D2 increases.
Case (b) If p1 increases, both D1 and D2 decreases. If p2 increases, both D1 and D2 decreases.
When the demand changes reversely, it can mean that one product is replaced by another: these might be competitive
products. For example, the same kind of product of two brands, like washing powder.
When the demand changes the same way, it can mean that the products go together: these might be supplementary
products. For example, hoover and dust bag.