Reduction of order

For a second-order homogeneous linear ODE

y ′′ + p(x)y ′ + q(x)y = 0, (1)

an initial value problem consists of (1) and two initial conditions

y(x0 ) = K0 , y ′ (x0 ) = K1 . (2)

These conditions prescribe given values of the solution and its first derivative at the same
point x0 .
The conditions determine the two arbitrary constants c1 and c2 in a general solution

y = c1 y1 + c2 y2 , (3)

where y1 and y2 are suitable solutions of the ODE.

This results in a unique solution passing through (x0 , K0 ) with slope K1 . This solution
is called the particular solution.

Example 1: Initial Value Problem

Solve:
y ′′ + y = 0, y(0) = 3.0, y ′ (0) = −0.5.
Solution.
The functions cos x and sin x are solutions. Hence,

y = c1 cos x + c2 sin x.

Differentiate:
y ′ = −c1 sin x + c2 cos x.
Apply initial conditions:
y(0) = c1 = 3.0,
y ′ (0) = c2 = −0.5.
Thus,
y = 3.0 cos x − 0.5 sin x.

1
1 Definition: General Solution, Basis, Particular Solu-
tion
A general solution of the ODE is

y = c1 y1 + c2 y2 ,

where y1 and y2 are solutions that are not proportional, and c1 , c2 are arbitrary constants.
The pair (y1 , y2 ) is called a basis (or fundamental system).
A particular solution is obtained by assigning specific values to c1 and c2 .

2 Linear Independence
Two functions y1 , y2 are called linearly independent on an interval I if

k1 y1 (x) + k2 y2 (x) = 0 ∀x ∈ I

implies
k1 = 0, k2 = 0.
They are linearly dependent if they are proportional:

y1 = ky2 or y2 = ly1 .

3 Definition: Basis (Reformulated)

A basis of solutions is a pair of linearly independent solutions of the ODE.

4 Reduction of Order
If one solution y1 of
y ′′ + p(x)y ′ + q(x)y = 0
is known, a second solution can be obtained by assuming

y = u(x)y1 (x).

Example 1: Reduction of Order

Solve:
(x2 − x)y ′′ − xy ′ + y = 0, y1 = x.
Solution.
Let:
y = ux.

2
 Then:
y ′ = u′ x + u, y ′′ = u′′ x + 2u′ .

(x2 − x)(u′′ x + 2u′ ) − x(u′ x + u) + ux = 0.

(x2 − x)u′′ x + (2x2 − 2x − x2 )u′ = 0.

(x2 − x)u′′ + (x − 2)u′ = 0.

Let v = u′ :
(x2 − x)v ′ + (x − 2)v = 0.

dv x−2
=− 2 dx.
v x −x

ln |v| = ln |x − 1| − 2 ln |x|.

x−1
v= .
x2
1
u = ln |x| + .
x

y2 = x ln |x| + 1.

y1 = x, y2 = x ln |x| + 1.

Example 2
1
x2 y ′′ + 4xy ′ + 2y = 0, y1 =
x
u
y = u(x) y1 =
x
u′′u
y = − 2
x x
u′′ 2u′ 2u
y ′′ = − 2 + 3
x x x
2u
x2 y ′′ = xu′′ − 2u′ +
x
4u
4xy ′ = 4u′ −
x
3
 2u
2y =
x
2u 4u 2u
xu′′ − 2u′ + + 4u′ − + =0
x x x

xu′′ + 2u′ = 0

2
u′′ + u′ = 0
x
Let v = u′
2
v′ + v = 0
x
dv 2
= − dx
v x

ln |v| = −2 ln x

1
v=
x2
Z
1 1
u= 2
dx = −
x x
u 1
y2 = =− 2
x x
Exercise Problems:

1.
x2 y ′′ − 5xy ′ + 9y = 0, y1 = x3

2.
cos x
xy ′′ + 2y ′ + xy = 0, y1 =
x

4
 Wronskian, Linear Independence
Dr. Sk Shahid Nadim

1 Linear Independence and Dependence

Definition
Two functions y1 (x) and y2 (x) are said to be linearly independent on an interval I
if
k1 y1 (x) + k2 y2 (x) = 0 ∀x ∈ I
implies
k1 = 0, k2 = 0.

Definition
Two functions y1 (x) and y2 (x) are said to be linearly dependent on an interval I if
there exist constants k1 and k2 , not both zero, such that

k1 y1 (x) + k2 y2 (x) = 0 ∀x ∈ I.

Important Idea
If two functions are linearly dependent, then one function is a constant multiple of the
other:
y1 (x) = k y2 (x)
Interpretation:

• Both functions represent the same curve (just scaled).

• They do not provide new information.

• Hence, they cannot form a general solution.

1
2 For n functions
Definition
A set of functions {y1 , y2 , . . . , yn } is said to be linearly independent if

c1 y1 + c2 y2 + · · · + cn yn = 0

implies that
c1 = c2 = · · · = cn = 0.
Meaning: No function in the set can be written as a linear combination of the others.

3 Wronskian (For two functions)

Definition
The Wronskian of two functions y1 (x) and y2 (x) is defined as:

y1 (x) y2 (x)
W (y1 , y2 )(x) = = y1 y2′ − y2 y1′
y1′ (x) y2′ (x)

Interpretation
• The Wronskian is a determinant measuring the variation between functions.

• W = 0 ⇒ no variation ⇒ same direction ⇒ dependent

• W ̸= 0 ⇒ variation exists ⇒ independent

4 Wronskian (For n functions)

Definition
The Wronskian of n functions y1 (x), y2 (x), . . . , yn (x) is defined as:

y1 (x) y2 (x) ··· yn (x)

y1′ (x) y2′ (x) ··· yn′ (x)
W (y1 , y2 , . . . , yn )(x) = y1′′ (x) y2′′ (x) ··· yn′′ (x)
.. .. ... ..
. . .
(n−1) (n−1) (n−1)
y1 (x) y2 (x) · · · yn (x)

2
5 Wronskian Theorem (For two variable)
Theorem
Let y1 (x) and y2 (x) be two solutions of a linear differential equation on an interval I.

• If W (y1 , y2 )(x0 ) = 0 at some point x0 , then y1 and y2 are linearly dependent on

the interval.

• If W (y1 , y2 )(x) ̸= 0 at some point, then y1 and y2 are linearly independent.

6 Wronskian Theorem (General Form)

Theorem
Let y1 (x), y2 (x), . . . , yn (x) be n solutions of an n-th order linear differential equation
on an interval I.
Then:

• If
W (y1 , y2 , . . . , yn )(x0 ) ̸= 0
for some x0 ∈ I, then the functions y1 , y2 , . . . , yn are linearly independent on I.

• If
W (y1 , y2 , . . . , yn )(x) = 0 for all x ∈ I,
then the functions are linearly dependent on I.

7 Solved Examples (Wronskian Method)

Method
To determine whether two functions are linearly independent or dependent:

1. Compute derivatives y1′ , y2′

2. Form Wronskian:
W = y1 y2′ − y2 y1′

3. Simplify

4. Decision:

• W ̸= 0 ⇒ Independent
• W = 0 ⇒ Dependent

3
Question
Check whether y1 = ex and y2 = xex are linearly independent.

Solution
Step 1: Compute derivatives
y1′ = ex
y2′ = (1 + x)ex
Step 2: Form Wronskian

ex xex
W =
ex (1 + x)ex

Step 3: Evaluate
W = ex (1 + x)ex − xex ex = e2x
Step 4: Conclusion
W ̸= 0 ⇒ Linearly Independent

Question
Check whether y1 = ex and y2 = 5ex are linearly independent.

Solution

y1′ = ex , y2′ = 5ex

ex 5ex
W = =0
ex 5ex

W = 0 ⇒ Linearly Dependent

4
Question
Check whether y1 = cos x and y2 = sin x are linearly independent.

Solution

y1′ = − sin x, y2′ = cos x

cos x sin x
W =
− sin x cos x

W = cos2 x + sin2 x = 1

W ̸= 0 ⇒ Linearly Independent

Question

Check whether y1 = x2 and y2 = x3 are linearly independent.

Solution

y1′ = 2x, y2′ = 3x2

x2 x3
W =
2x 3x2

W = 3x4 − 2x4 = x4

W ̸= 0 ⇒ Linearly Independent

5
 Question

Check whether y1 = x2 and y2 = 3x2 are linearly independent.

Solution

y1′ = 2x, y2′ = 6x

x2 3x2
W =
2x 6x

W = 6x3 − 6x3 = 0

W = 0 ⇒ Linearly Dependent

Question

Check whether y1 = e2x and y2 = e3x are linearly independent.

Solution

y1′ = 2e2x , y2′ = 3e3x

e2x e3x
W =
2e2x 3e3x

W = 3e5x − 2e5x = e5x

W ̸= 0 ⇒ Linearly Independent

Final Result
• W ̸= 0 ⇒ Independent

• W = 0 ⇒ Dependent

6
Homogeneous linear Equation with Constant Co-efficients

1 Let us consider a homogeneous linear equation of degree 2

d2 y dy
2
+a + by = 0,
dx dx
where a and b are constants.
2 Substituting y = e λx in the above equation, we get the following
quadratic equation:

λ2 + aλ + b = 0.
This is called auxiliary equation.
3 There are two roots of the above equation namely,
√ √
−a + a2 − 4b −a − a2 − 4b
λ1 = , and λ2 =
2 2

April 6, 2026 1 / 11
Homogeneous linear Equation with Constant Co-efficients

There are three different scenarios, which are as follows.

1 a2 − 4b > 0. Then λ1 and λ2 are two real and distinct roots.
−a
2 a2 − 4b = 0. Then λ1 = λ2 = λ = 2 . Hence, in this case, we get
two equal real roots.
3 a2 − 4b < 0. In this case, we get two complex conjugate roots

−a + iω −a − iω
λ1 = and λ2 =
2 2
a2
, where ω 2 = b − 4.

April 6, 2026 2 / 11
Two distinct real roots

1 At first, we shall consider the case where λ1 and λ2 are real and
distinct.
2 The General Solution of the differential equation looks like

y = (C1 e λ1 x + C2 e λ2 x ).

April 6, 2026 3 / 11
Example of the case with two distinct real roots

Let us take the following differential equation:

d2 y dy
2
+ − 6y = 0. y (0) = 10, y ′ (0) = 0.
dx dx
▶ The corresponding auxiliary equation is

λ2 + λ − 6 = 0.
▶ Roots are λ1 = −3, λ2 = 2.
▶ The general solution is

y = (C1 e −3x + C2 e 2x ).

April 6, 2026 4 / 11
Example of the case with two distinct real roots (Continued)

▶
y (0) = 10 =⇒ C1 + C2 = 10.
▶
y ′ (0) = 0 =⇒ 2C2 − 3C1 = 0.
▶ Solving the above equations, we get

C1 = 4, C2 = 6.

▶ Hence, the required general solution is

y = (4e −3x + 6e 2x ).

April 6, 2026 5 / 11
Two equal roots

1 We shall now consider the case where λ1 and λ2 are real and equal.
2 Furthermore, we assume λ1 = λ2 = λ.
3 The General Solution of the differential equation is

y = e λx (C1 + C2 x).

April 6, 2026 6 / 11
Example of the case with two equal (real) roots

▶ Let us consider the following differential equation:

d2 y dy
+4 + 4y = 0. y (0) = 4, y ′ (0) = 6.
dx 2 dx
▶ The corresponding auxiliary equation is

λ2 + 4λ + 4 = 0.

▶ Roots are λ1 = λ2 = λ = −2.

▶ The general solution is

y = e −2x (C1 + C2 x)

April 6, 2026 7 / 11
Example of the case with two equal (real) roots (Continued)

▶
y (0) = 4 =⇒ C1 = 4.
▶
y ′ (0) = 6 =⇒ −2C1 + C2 = 6.
▶ Solving the above equations, we get

C1 = 4, C2 = 14.

▶ Hence, the required general solution is

y = e −2x (4 + 14x)

April 6, 2026 8 / 11
Two complex Roots

1 We shall now consider the case where λ1 and λ2 are complex.

2 The general solution of the differential equation is

y = e νx (C1 cos wx + C2 sin wx),

, if λ1 = ν + iω and λ2 = ν − iω are roots of the auxiliary equation.

April 6, 2026 9 / 11
Example of the case with two complex roots

▶ Let us consider the following differential equation:

d2 y dy 1
2 2
+6 + 5y = 0. y (0) = 1, y ′ (0) = − .
dx dx 2
▶ The corresponding auxiliary equation is

5
λ2 + 3λ + = 0.
2
▶ Roots are λ1 = − 32 + i 12 and λ2 = − 32 − i 21 .
▶ Hence, here ν = − − 23 and ω = 12 .
▶ So, the general solution is
3x x x
y == e − 2 (C1 cos + C2 sin ).
2 2

April 6, 2026 10 / 11
Example of the case with two complex roots (Continued)

▶ y (0) = 1 =⇒ C1 = 1.
▶ y ′ (0) = − 12 =⇒ C2 − 3C1 = − 12 .
▶ Hence, we get C1 = 1, C2 = 5.
▶ So, the required general solution is
3x x x
y = e − 2 (5 cos + sin ).
2 2

April 6, 2026 11 / 11
 Variation of Parameters

Variation of Parameters is a method used to find particular solutions of non-homogeneous

linear differential equations.

Second-Order Linear Differential Equation

We consider the second-order linear differential equation

y ′′ + p(x)y ′ + q(x)y = g(x), (1)

where p(x), q(x), and g(x) are continuous functions on some interval.
The following steps are essential for the method ‘Variation of Parameters’:

• Step 1: First solve the corresponding homogeneous equation.

• Step 2: Replace the arbitrary constants in the complementary solution with some
functions of independent variables.

• Step 3: Determine these functions following some steps so that the new expression
becomes a particular solution of the given non-homogeneous equation.

Step 1: Solve the Homogeneous Equation

Consider the associated homogeneous equation

y ′′ + p(x)y ′ + q(x)y = 0.

Let y1 (x) and y2 (x) be two linearly independent solutions of this homogeneous equa-
tion.
The general solution (complementary solution) of the equation is

yc = c1 y1 (x) + c2 y2 (x), (*)

where c1 and c2 are arbitrary constants.

1
Step 2: Choose a Particular Solution
Consider a particular solution to solve the non-homogeneous equation as follows:

yp = u1 (x)y1 (x) + u2 (x)y2 (x),

where u1 (x) and u2 (x) need to be determined.

Step 3: Differentiate yp to obtain yp′ and yp′′

yp = u1 (x)y1 (x) + u2 (x)y2 (x)

Differentiating [Link] x gives

yp′ = u′1 (x)y1 (x) + u1 (x)y1′ (x) + u′2 (x)y2 (x) + u2 (x)y2′ (x).

To simplify the expression, choose the auxiliary condition

u′1 (x)y1 (x) + u′2 (x)y2 (x) = 0. (2)

Then we have
yp′ = u1 (x)y1′ (x) + u2 (x)y2′ (x).

Now differentiating again [Link] x gives

yp′′ = u′1 (x)y1′ (x) + u1 (x)y1′′ (x) + u′2 (x)y2′ (x) + u2 (x)y2′′ (x).

Step 4: Substitute yp′ and yp′′ into Main Equation

Substitute yp , yp′ , and yp′′ into the non-homogeneous equation

y ′′ + p(x)y ′ + q(x)y = g(x)

we have
 ′ ′
u1 y1 + u1 y1′′ + u′2 y2′ + u2 y2′′ + p(x) u1 y1′ + u2 y2′ + q(x) u1 y1 + u2 y2 = g(x)
    

⇒ (u′1 y1′ + u′2 y2′ ) + u1 y1′′ + py1′ + qy1 + u2 y2′′ + py2′ + qy2 = g(x).

Now, y1 and y2 are solutions of the homogeneous equation,

y1′′ + py1′ + qy1 = 0, y2′′ + py2′ + qy2 = 0.

So we have
u′1 (x)y1′ (x) + u′2 (x)y2′ (x) = g(x). (3)

2
Step 5: Obtain the System for u′1 and u′2
From (2) and (3), we now have the following two equations:

u′1 (x)y1 (x) + u′2 (x)y2 (x) = 0,

u′1 (x)y1′ (x) + u′2 (x)y2′ (x) = g(x).

Step 6: Solve the System Using the Wronskian

The determinant of the system is the Wronskian:

y1 (x) y2 (x)
W (y1 , y2 )(x) = ′ ′
= y1 (x)y2′ (x) − y2 (x)y1′ (x).
y1 (x) y2 (x)

By Cramer’s rule, we have

0 y2 (x)
g(x) y2′ (x) −y2 (x)g(x)
u′1 (x) = = ,
W (y1 , y2 ) W (y1 , y2 )

y1 (x) 0
y1′ (x) g(x) y1 (x)g(x)
u′2 (x) = = .
W (y1 , y2 ) W (y1 , y2 )

Step 7: Integrate to Find u1 and u2

Direct integration gives

−y2 (x)g(x)
Z Z
y1 (x)g(x)
u1 (x) = dx, u2 (x) = dx.
W (y1 , y2 ) W (y1 , y2 )

Substituting the values into

yp = u1 (x)y1 (x) + u2 (x)y2 (x). (**)

gives the particular solution.

Step 8: General Solution

The general solution of the given non-homogeneous differential equation is

y = yc + yp = c1 y1 (x) + c2 y2 (x) + yp .

3
Example 1. By the method of variation of parameters, solve

y ′′ − y = ex

Solution 1. The homogeneous equation is

y ′′ − y = 0.

The auxiliary equation is

m2 − 1 = 0 ⇒ m = −1, 1.

Hence, two linearly independent solutions are

y1 (x) = ex , y2 (x) = e−x .

Therefore, the complementary solution is

yc = c1 ex + c2 e−x .

Consider the particular solution as

yp = u1 (x)ex + u2 (x)e−x .

Now, the Wronskian of y1 and y2 is

ex e−x
W (y1 , y2 ) = = −1 − 1 = −2.
ex −e−x

Here g(x) = ex . So,

−y2 (x)g(x) −e−x · ex 1

u′1 = = = ,
W (y1 , y2 ) −2 2

and
y1 (x)g(x) ex · e x e2x
u′2 = = =− .
W (y1 , y2 ) −2 2
Hence, Z
1 x
u1 = dx = ,
2 2
and
e2x 1 e2x e2x
Z
u2 = − dx = − · =− .
2 2 2 4
Henceforth, the particular solution is

4
 x  2x 
e x 1
yp = u1 y1 + u2 y2 = x
e + − e−x = ex − ex .
2 4 2 4

So, the general solution is

x 1
y = yc + yp = C1 ex + C2 e−x + ex − ex .
2 4
Example 2. Find a general solution to the following differential equation

2y ′′ + 18y = 6 tan(3t)

Solution 2. The homogeneous equation is

y ′′ + 9y = 0.

The auxiliary equation is

m2 + 9 = 0 ⇒ m = −3i, 3i.

Hence, two linearly independent solutions are

y1 (x) = cos(3t), y2 (x) = sin(3t).

Therefore, the complementary solution is

yc (t) = c1 cos(3t) + c2 sin(3t).

Consider the particular solution as

yp (t) = u1 (t) cos(3t) + u2 (t) sin(3t).

Now, the Wronskian of y1 and y2 is

cos(3t) sin(3t)
W (y1 , y2 ) = = 3(cos2 (3t) + sin2 (3t)) = 3.
−3 sin(3t) 3 cos(3t)

Here g(x) = 3 tan(3t). So,

−y2 (t)g(t) −3 tan(3t) sin(3t) − sin2 (3t)

u′1 (t) = = = ,
W (y1 , y2 ) 3 cos(3t)

and
y1 (t)g(t) 3 tan(3t) cos(3t)
u′2 (t) = = = sin(3t).
W (y1 , y2 ) 3

5
 Hence,

sin2 (3t)
Z Z  
1 1
u1 = − dt = − [sec(3t)−cos(3t)] dt = − ln | sec(3t) + tan(3t)| − sin(3t) ,
cos(3t) 3 3

and Z
1
u2 = sin(3t) dt = − cos(3t).
3
Henceforth, the particular solution is

cos(3t) sin(3t) cos(3t)

yp = u1 y1 + u2 y2 = − [ln | sec(3t) + tan(3t)| − sin(3t)] −
3 3
cos(3t)
=− ln | sec(3t) + tan(3t)|.
3
So, the general solution is

cos(3t)
y = yc + yp = c1 cos(3t) + c2 sin(3t) − ln | sec(3t) + tan(3t)|.
3

Flow-chart of the method:

Solve homogeneous equation −→ Complementary solution (yc ) −
→ Assume yp = u1 y1 + u2 y2
↓
Particular solution (yp ) ←
− Integrate to get u1 , u2 ←
− Find u′1 , u′2
↓
y = y c + yp

Exercise 1. Using method of variation of parameters, find a general solution to the

following differential equations
et
1. y ′′ − 2y ′ + y = t2 +1
;

2. y ′′ + 3y ′ + 2y = 1
1+ex

3. y ′′ + y = x sin x

4. y ′′ − 2y ′ + 2 = ex tan x

Exercise 2. Find the general solution to

ty ′′ − (t + 1)y ′ + y = t2

given that
y1 (t) = et , y2 (t) = t + 1

form a fundamental set of solutions for the homogeneous differential equation.

6
Exercise 3. Find the general solution to
9
x2 y ′′ − 2xy ′ + 2y = x 2

given that
y1 (x) = x, y2 (x) = x2

are solutions of the complementary equation x2 y ′′ − 2xy ′ + 2y = 0.

Exercise 4. Find the general solution to

(x − 1)y ′′ − xy ′ + y = (x − 1)2

given that
y1 (x) = x, y2 (x) = ex

are solutions of the complementary equation (x − 1)y ′′ − xy ′ + y = 0.