Review of Power Series

Solutions about Ordinary Points

Solutions about Singular Points
Summary

Chapter 6: Series Solutions of Linear Equations

王奕翔
Department of Electrical Engineering
National Taiwan University

[email protected]

November 13, 2013

1 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Solving Higher-Order Linear Equations

In Chapter 4, we learns how to analytically solve two special kinds of
higher-order linear differential equations:
1 Linear Differential Equation with Constant Coefficients
2 Cauchy-Euler Equations

Essentially only one kind – linear DE with constant coefficients!

Because to solve Cauchy-Euler DE, we substitute x = et !

Question: Is it possible to solve other kinds, like the following?

(x2 + 2x − 3)y′′ − 2(x + 1)y′ + 2y = 0

2 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Idea: Express the solution function as a

power series!
∞
∑
y(x) = cn xn
n=0

3 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Focus on: Linear Second-Order Differential Equations

Throughout this lecture, we shall focus on solving homogeneous linear
second order differential equations

a2 (x)y′′ + a1 (x)y + a0 (x) = 0,

using the method of power series.

Standard Form: Frequently throughout the discussions in this lecture:

y′′ + P(x)y′ + Q(x)y = 0.

4 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1 Review of Power Series

2 Solutions about Ordinary Points

3 Solutions about Singular Points

4 Summary

5 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Power Series

Definition
A power series in (x − a) (or a power series centered at a) is an infinite
series of the following form:
∞
∑
cn (x − a)n ,
n=0

where {cn }∞
0 is a sequence of real numbers.

Some Examples:
∞
∑ ∞
∑
n
x = 1 + x + x + ··· ,
2
2n xn = 1 + 2x + 4x2 + · · · .
n=0 n=0

6 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Convergence, Divergence, Absolute Convergence

∑∞
Convergence: A power series n=0 cn (x − a)n converges at x = x0 if

∑
N
lim cn (x0 − a)n exists.
N→∞
n=0

Otherwise, the power series diverges at x = x0 .

∑
Absolute Convergence: A power series ∞ n
n=0 cn (x − a) converges absolutely
at x = x0 if
∑N
lim |cn (x0 − a)n | exists.
N→∞
n=0

Ratio Test: Suppose cn ̸= 0 for all n, then the following test tells us about the
convergence of the series:
    <1 absolute convergence
 cn+1 (x0 − a)n+1   cn+1 
lim   
= |x0 − a| lim   >1 divergence
n→∞  cn (x0 − a)n  n→∞ cn 
=1 not sure

7 / 43 王奕翔 DE Lecture 9
 is a power series in x.
Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary Important Facts T
facts about power series !#n
Interval of Convergence • Convergence A p
its sequence of par
Interval of Convergence: Every power series has an interval ofNlim !Nn!0 cn (x " a
:#
convergence I = (a − R, a + R), in which he power series is said to be diverg
∑∞ • Interval of Converg
n=0 cn (x − a) converges absolutely.
n
The interval of conve
R > 0 is called the radius of convergence. converges. The cente
• Radius of Converge
absolute power series is called
divergence convergence divergence converges for " x " a
x converges only at its
a−R a a+R
we write R ! #. Re
series may equivalent to the sim
converge or diverge may or may not conv
at endpoints
• Absolute Converg
FIGURE 6.1.1 Absolute convergence converges absolute
∞
within the interval of convergence∑ and and is not an endpo
A power series defines a function
divergence ofofx,this
outside f(x) :=
interval cn (x − a)n for!#n!0
x ∈" cI.n(x " a)n " co
n=0

8 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

∑∞
Function Defined by a Power Series n=0 cn (x − a)n
Define the function (I: interval of convergence)
∞
∑
y(x) := cn (x − a)n , x ∈ I.
n=0

Differentiation
∞
∑
y′ (x) = c1 + 2c2 x + 3c3 x2 + · · · = ncn (x − a)n−1 , x ∈ I
n=1
∑∞
y′′ (x) = 2c2 + 6c3 x + 12c4 x2 + · · · = n(n − 1)cn (x − a)n−2 , x ∈ I
n=2

9 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Taylor’s Series

If a function f(x) is infinitely differentiable at a point a, then it can be

represented by Taylor’s Series as follows, with a radius of convergence
R > 0.
∑∞ (n)
f (a)
f(x) = (x − a)n .
n=0
n!
Examples
∑∞
x x2 xn
ex = 1 + + + ··· = , x∈R
1! 2! n=0
n!
∑∞
1
= 1 + x + x2 + · · · = xn , x ∈ (−1, 1)
1−x n=0

10 / 43 王奕翔 DE Lecture 9
 Review of Power Series
$ (n)
(0) n Points
f Ordinary f%(0) f &(0) 2 . . .
!
Solutions about
x #Points
Solutions about Singular f(0) ! x! x ! .
n#0 n! Summary 1! 1!
You might remember some of the following Maclaurin series representations.

Interval
Maclaurin Series of Convergence

x x2 x3 $
1
ex # 1 ! ! ! ! . . . # ! xn ("$, $)
1! 2! 3! n#0 n!

x2 x4 x6 . . . $
("1)n 2n
cos x # 1 " ! " ! # ! x ("$, $)
2! 4! 6! n#0 (2n)!

x3 x5 x7 . . . $
("1)n 2n!1
sin x # x " ! " ! # ! x ("$, $)
3! 5! 7! n#0 (2n ! 1)!

x3 x5 x7 . . . $
("1)n 2n!1
tan"1 x # x " ! " ! # ! x ["1, 1] (2)
3 5 7 n#0 2n ! 1

x2 x4 x6 . . . $
1 2n
cosh x # 1 ! ! ! ! # ! x ("$, $)
2! 4! 6! n#0 (2n)!

x3 x5 x7 . . . $
1
sinh x # x ! ! ! ! # ! x2n!1 ("$, $)
3! 5! 7! n#0 (2n ! 1)!

2 3 4 n!1
x x x ("1) $
ln(1 ! x) # x " ! " !...# ! xn ("1, 1]
2 3 4 n#1 n
1 $
# 1 ! x ! x2 ! x3 ! . . . # ! xn ("1, 1)
1"x n#0

These results can be used to obtain power series representations of other

functions. For example, if we wish to find the Maclaurin series representation
2
of, say, ex we need only replace x in the Maclaurin series for ex:
11 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1 Review of Power Series

2 Solutions about Ordinary Points

3 Solutions about Singular Points

4 Summary

12 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Ordinary and Singular Points

Focus on homogeneous linear 2nd order DE

a2 (x)y′′ + a1 (x)y′ + a0 (x)y = 0

Rewrite it into its standard form y′′ + P(x)y′ + Q(x)y = 0 .

Definition (Ordinary and Singular Points)

x = x0 is an ordinary point of the above DE if both P(x) and Q(x) are
analytic at x0 . Otherwise, x = x0 is a singular point.

Analytic at a Point: a function f(x) is analytic ∑

at a point x = x0 if and
∞
only if f(x) can be represented as a power series n=0 cn (x − x0 )n with a
positive radius of convergence.
In our lecture analytic ≡ infinitely differentiable.

13 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Examples: Ordinary and Singular Points

1 Constant coefficients: a2 y′′ + a1 y′ + a0 y = 0. Every x ∈ R is ordinary.

2 Cauchy-Euler DE: x2 y′′ + xy′ + y = 0.

P(x) = 1x is analytic at x ∈ R \ {0}.
Q(x) = x12 is analytic at x ∈ R \ {0}.
Hence, x = 0 is the only singular point.

3 Polynomial Coefficients: a2 (x)y′′ + a1 (x)y′ + a0 (x)y = 0, where

a2 (x) ̸= 0, a1 (x), a0 (x) are all polynomials of x.
a1 (x)
P(x) = a2 (x)
is analytic at x ∈ R \ {r ∈ R : a2 (r) = 0}.
a0 (x)
Q(x) = a2 (x)
is analytic at x ∈ R \ {r ∈ R : a2 (r) = 0}.
Hence, {r ∈ R : a2 (r) = 0} are singular points.

4 y′′ + xy′ + (ln x) y = 0.

P(x) = x is analytic at x ∈ R.
Q(x) = ln x is analytic at x ∈ (0, ∞).
Hence, every x ≤ 0 is singular.
14 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Existence of Power Series Solutions about Ordinary Points

The following theorem lays the theoretical foundations of the method.
Theorem
Let x = x0 be an ordinary point of a homogenous linear 2nd order DE.
Then, we can find two linearly independent solutions in the form of power
series centered at x0 , that is,
∞
∑
y= cn (x − x0 )n .
n=0

Moreover, the radius of convergence ≥ the distance from x0 to the

closest singular point in C.

15 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Example: Minimum Radius of Convergence

Example
Consider a linear second order DE (x2 + 1)y′′ + xy′ − y = 0.
Find the minimum radius of convergence of a power series solution about
the ordinary points x = −1 and x = 0.

A: The singular points in the complex domain C is ±i.

√ √
The distance between −1 and ±i is 12 + 12 = 2. The distance
between 0 and ±i is 1.
Based on the previous
√ theorem, we obtain the minimum radius of
convergence R = 2 and R = 1 respectively.
√
In other words, for |x + 1| < 2 and |x| < 1, the power series solution of
the DE exists (and converges absolutely).

16 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Example: Finding Power Series Solutions

Example
Consider a linear second order DE (x2 + 1)y′′ + xy′ − y = 0.
Find two linearly independent power series solution about the ordinary
point x = 0.
A: From the previous discussion, we know that the interval of definition
of the solutions should be (−1, 1).
∑∞
Plug in the power series representation y = n=0 cn xn :
∞
∑ ∞
∑ ∞
∑
n ′ n−1 ′′
y= cn x , y = ncn x , y = cn n(n − 1)xn−2
n=0 n=0 n=0
′′ ′
=⇒ 0 = (x + 1)y + xy − y
2

∑∞
{ 2 }
= (n − 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

17 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1
X
(x2 + 1)y 00 + xy 0 y (n2 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

18 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

y 00

1
X
(x2 + 1)y 00 + xy 0 y (n2 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

x2 y 00 + xy 0 y

19 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

y 00

1
X
(x2 + 1)y 00 + xy 0 y (n2 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

x2 y 00 + xy 0 y
1
X 1
X 1
X
2 00
x y =x 2
cn n(n 1)x n 2
xy 0 = x cn nxn 1
y= c n xn
n=0 n=0 n=0
1
X 1
X
= cn n(n 1)xn = cn nxn
n=0 n=0

20 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1
X
y 00 = cn n(n 1)xn 2
00
y n=0
1
X
= cn n(n 1)xn 2

n=2
X1
= ck+2 (k + 2)(k + 1)xk
k=0
1
X
(x2 + 1)y 00 + xy 0 y (n2 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

x2 y 00 + xy 0 y
1
X 1
X 1
X
2 00
x y =x 2
cn n(n 1)x n 2
xy 0 = x cn nxn 1
y= c n xn
n=0 n=0 n=0
1
X 1
X
= cn n(n 1)xn = cn nxn
n=0 n=0

21 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1
X
y 00 = cn n(n 1)xn 2
00
y n=0
1
X
= cn n(n 1)xn 2

n=2
X1
= ck+2 (k + 2)(k + 1)xk
k=0
1
X
(x2 + 1)y 00 + xy 0 y (n2 1)cn + (n + 2)(n + 1)cn+2 xn
n=0

x2 y 00 + xy 0 y
1
X 1
X 1
X
2 00
x y =x 2
cn n(n 1)x n 2
xy 0 = x cn nxn 1
y= c n xn
n=0 n=0 n=0
1
X 1
X
= cn n(n 1)xn = cn nxn
n=0 n=0

22 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Recursive Formula of Coefficients in Power Series Solution

Example
Consider a linear second order DE (x2 + 1)y′′ + xy′ − y = 0.
Find two linearly independent power series solution about the ordinary
point x = 0.
∑∞
Plug in the power series representation y = n=0 cn xn , we get
∞
∑ { 2 }
0= (n − 1)cn + (n + 2)(n + 1)cn+2 xn
n=0
=⇒ (n2 − 1)cn + (n + 2)(n + 1)cn+2 = 0, n = 0, 1, 2, . . .
1 1−n
=⇒ c2 = c0 , c3 = 0, cn+2 = cn , n = 2, 3, 4, . . .
2 2+n
1 −1 1·3
=⇒ c2 = c0 , c4 = c0 , c6 = c0 , · · ·
2 2·4 2·4·6
c3 = c5 = c7 = · · · = 0.
23 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

(n2 − 1)cn + (n + 2)(n + 1)cn+2 = 0, n ≥ 0

n=0
1
−c0 + 2c2 = 0 =⇒ c2 = c0
2
n=1
0 + 6c3 = 0 =⇒ c3 = 0
n≥2

(n − 1)(n + 1)cn + (n + 2)(n + 1)cn+2 = 0

1−n
=⇒ cn+2 = cn
2+n

24 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Wrapping Up

Example
Consider a linear second order DE (x2 + 1)y′′ + xy′ − y = 0.
Find two linearly independent power series solution about the ordinary
point x = 0.

Therefore
{ }
1 2 −1 4 1·3 6
y = c0 1 + x + x + x + · · · + c1 x
2 2·4 2·4·6

Thus we obtain two linearly independent solutions: y1 (x) = x, and

∑∞
1 1 · 3 · 5 · · · (2n − 3) 2n
y2 (x) = 1 + x2 + (−1)n−1 x , |x| < 1.
2 n=2
2n n!

25 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1 Review of Power Series

2 Solutions about Ordinary Points

3 Solutions about Singular Points

4 Summary

26 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Regular and Irregular Singular Points

Focus on homogeneous linear 2nd order DE
a2 (x)y′′ + a1 (x)y′ + a0 (x)y = 0

Rewrite it into its standard form y′′ + P(x)y′ + Q(x)y = 0 .

Definition (Regular and Irregular Singular Points)

A singular point x = x0 of the above DE is regular if both (x − x0 )P(x)
and (x − x0 )2 Q(x) are analytic at x0 . Otherwise, x = x0 is an irregular
singular point.

Note: There may not be power series solutions about a singular point
x = x0 . However, it is possible to obtain a generalized power series
solution
∞
∑
r
y(x) = (x − x0 ) cn (x − x0 )n
n=0

27 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

In other words, at a regular singular point x = x0 , we can convert the

standard form
y′′ + P(x)y′ + Q(x)y = 0,
into
(x − x0 )2 y′′ + (x − x0 )p(x)y′ + q(x)y = 0 .

where p(x) = (x − x0 )P(x) and q(x) = (x − x0 )2 Q(x) are both analytic at

x = x0 , that is,
∞
∑ ∞
∑
p(x) = an (x − x0 )n , q(x) = bn (x − x0 )n , |x − x0 | < R
n=0 n=0

for some R > 0.

28 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Examples: Classification of Singular Points

1 Cauchy-Euler DE: x2 y′′ + xy′ + y = 0. Its has one singular point x = 0.
xP(x) = x 1x = 1 is analytic at x = 0.
x2 Q(x) = x2 x12 = 1 is analytic at x = 0.
Hence, x = 0 is a regular singular point.

2 Polynomial Coefficients: a2 (x)y′′ + a1 (x)y′ + a0 (x)y = 0, where

a2 (x) ̸= 0, a1 (x), a0 (x) are all polynomials of x. Let x = x0 be a root of
a2 (x) = 0. Hence x = x0 is a singular point.
If in the denominator of the rational function P(x) = aa12 (x)
(x)
(after
reduction), the factor (x − x0 ) appears at most to the first power,
then (x − x0 )P(x) is analytic at x = x0 .

If in the denominator of the rational function Q(x) = aa02 (x)

(x)
(after
reduction), the factor (x − x0 ) appears at most to the second power,
then (x − x0 )2 Q(x) is analytic at x = x0 .

29 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Examples: Classification of Singular Points

Example
For the second order DE (x2 − 4)2 y′′ + 3(x − 2)y′ + 5y = 0, find the singular
points and classify them into regular and irregular ones.

A: First rewrite the DE into the standard form:

x−2 ′ 5
y′′ + 3 y + 2 y = y′′ + P(x)y′ + Q(x)y = 0.
(x2 − 4)2 (x − 4)2

Since P(x) = 3 (x2x−2

−4)2
3
= (x−2)(x+2)2 and Q(x) =
5
(x2 −4)2
= 5
(x−2)2 (x+2)2
, we
have two singular points x = 2, −2 for this DE.
x = 2: regular singular point, because (x − 2)P(x) = 3
(x+2)2
and
(x − 2)2 Q(x) = (x+2)
5
2 are both analytic at x = 2.

x = −2: irregular singular point because (x + 2)P(x) = 3

(x−2)(x+2)
is not
analytic at x = −2.

30 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Method of Frobenius

Theorem
Let x = x0 be a regular singular point of a homogenous linear 2nd order
DE. Then, we can find at least one solutions in the following form:
∞
∑ ∞
∑
y = (x − x0 )r cn (x − x0 )n = cn (x − x0 )n+r ,
n=0 n=0

where r is a constant (not necessarily an integer) to be determined. The

series will converge on some interval 0 < x − x0 < R.

Note 1: Without loss of generality we assume that c0 ̸= 0.

Note 2: We have to determine
The exponent r first,
and then the sequence {cn , n = 1, 2, . . .}.

31 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Method of Frobenius: Calculation

Without loss of generality, assume that the regular singular point is x = 0. We
convert the standard form into

x2 y′′ + xp(x)y′ + q(x)y = 0

∑ ∑∞
where p(x) = xP(x) = ∞ n 2
n=0 an x and q(x) = x Q(x) =
n
n=0 bn x .
∑ ∑∞
Plug in y = xr ∞ n
n=0 cn x = n=0 cn x
n+r
, we get
(∞ )
2 ′′ ′
∑ n+r−2
x y + xp(x)y + q(x)y = x 2
cn (n + r)(n + r − 1)x
n=0
( )( )
∑
∞
n
∑
∞
n+r−1
+x an x cn (n + r)x
n=0 n=0
( )( )
∑
∞
n
∑
∞
n+r
+ bn x cn x
n=0 n=0

32 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Method of Frobenius: Calculation

Without loss of generality, assume that the regular singular point is x = 0. We
convert the standard form into

x2 y′′ + xp(x)y′ + q(x)y = 0

∑ ∑∞
where p(x) = xP(x) = ∞ n 2
n=0 an x and q(x) = x Q(x) =
n
n=0 bn x .
∑ ∑∞
Plug in y = xr ∞ n
n=0 cn x = n=0 cn x
n+r
, we get
(∞ )
2 ′′ ′ r
∑ n
x y + xp(x)y + q(x)y = x cn (n + r)(n + r − 1)x
n=0
(∞ )( ∞ )
r
∑ n
∑ n
+x an x cn (n + r)x
n=0 n=0
(∞ )( )
r
∑ n
∑
∞
n
+x bn x cn x
n=0 n=0

33 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Method of Frobenius: Calculation

Without loss of generality, assume that the regular singular point is x = 0. We
convert the standard form into

x2 y′′ + xp(x)y′ + q(x)y = 0

∑ ∑∞
where p(x) = xP(x) = ∞ n 2
n=0 an x and q(x) = x Q(x) =
n
n=0 bn x .
∑ ∑∞
Plug in y = xr ∞ n
n=0 cn x = n=0 cn x
n+r
, we get
(∞ )
2 ′′ ′ r
∑ n
x y + xp(x)y + q(x)y = x cn (n + r)(n + r − 1)x
n=0
( { n } )
r
∑
∞ ∑ n
+x an−k ck (k + r) x
n=0 k=0
( { n } )
r
∑
∞ ∑ n
+x bn−k ck x
n=0 k=0

34 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Method of Frobenius: Calculation

Without loss of generality, assume that the regular singular point is x = 0. We
convert the standard form into

x2 y′′ + xp(x)y′ + q(x)y = 0

∑ ∑∞
where p(x) = xP(x) = ∞ n 2
n=0 an x and q(x) = x Q(x) =
n
n=0 bn x .
∑ ∑∞
Plug in y = xr ∞ n
n=0 cn x = n=0 cn x
n+r
, we get
(∞ )
2 ′′ ′ r
∑ n
x y + xp(x)y + q(x)y = x Ln x
n=0

where

∑
n
Ln := cn (n + r)(n + r − 1) + ck {an−k (k + r) + bn−k } = 0, ∀ n = 0, 1, 2, . . .
k=0

35 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Indicial Equation (Index → Indices → Indicial)

Further manipulate the conditions:

∑
n
Ln = cn (n + r)(n + r − 1) + ck {an−k (k + r) + bn−k }
k=0

∑
n−1
= cn {(n + r)(n + r − 1) + a0 (n + r) + b0 } + ck {an−k (k + r) + bn−k }
k=0

∑
n−1
= cn I (n + r) + ck {an−k (k + r) + bn−k } = 0.
k=0

For n = 0, the condition reduces to

I (r) = r(r − 1) + a0 r + b0 = 0 .

This is called the indicial equation of the problem, and the two roots are
called indicial roots/exponents.

36 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Roots of the Indicial Equation

Let the two real roots of I(r) = r(r − 1) + a0 r + b0 = 0 be r1 , r2 and r1 ≥ r2 .
P.S. We do not consider the case when r1 , r2 are complex conjugate roots.

1 r1 > r2 and r1 − r2 ∈
/ Z: Two linearly independent solutions can be found:
∑
∞ ∑
∞
y1 (x) = cn xn+r1 , c0 ̸= 0, y2 (x) = dn xn+r2 , d0 ̸= 0
n=0 n=0

2 r1 > r2 and r1 − r2 ∈ Z: Two linearly independent solutions can be found:

∑
∞ ∑
∞
y1 (x) = cn xn+r1 , c0 ̸= 0, y2 (x) = |{z}
C y1 (x) ln x+ dn xn+r2 , d0 ̸= 0.
n=0 can be 0 n=0

3 r1 = r2 : Two linearly independent solutions can be found:

∑
∞ ∑
∞
y1 (x) = cn xn+r1 , c0 ̸= 0, y2 (x) = y1 (x) ln x + dn xn+r2 .
n=0 n=1

37 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Examples

Example
Solve 2xy′′ + (1 + x)y′ + y = 0.

Example
Solve xy′′ + y = 0.

38 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

1 Review of Power Series

2 Solutions about Ordinary Points

3 Solutions about Singular Points

4 Summary

39 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Solve a2 (x)y 00 + a1 (x)y 0 + a0 (x)y = 0

about a point x = x0

Convert it into
y 00 + P (x)y 0 + Q(x)y = 0

P (x), Q(x) analytical at x0 ?

x0 is ordinary

PlugP
in
Yes 1
y = n=0 cn (x x0 ) n

Two linearly independent

power series solutions

40 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Solve a2 (x)y 00 + a1 (x)y 0 + a0 (x)y = 0 Convert it into

about a point x = x0 (x x0 )2 y 00 + (x x0 )p(x)y 0 + q(x)y = 0

p(x), q(x) analytical at x0 ?

Convert it into
y 00 + P (x)y 0 + Q(x)y = 0 x0 is regular singular
No
P (x), Q(x) analytical at x0 ?
Plug in P1
Yes y = (x x0 )r n=0 cn (x x0 ) n

x0 is ordinary Indicial Equation

r(r 1) + a0 r + b0 = 0
u := x x0
Case 1: r1 > r2 and r1 /Z
r2 2
PlugP
in
Yes 1
y = n=0 cn (x x0 ) n
1
X 1
X
y1 (x) = cn un+r1 , c0 6= 0, y2 (x) = dn un+r2 , d0 6= 0
n=0 n=0
Case 2: r1 > r2 and r1 r2 2 Z
Two linearly independent
1
X 1
X
power series solutions
y1 (x) = cn un+r1 , c0 6= 0, y2 (x) = |{z}
C y1 (x) ln u + dn un+r2 , d0 6= 0.
n=0 can be 0 n=0

Case 3: r1 = r2
1
X 1
X
y1 (x) = cn un+r1 , c0 6= 0, y2 (x) = y1 (x) ln u + dn un+r2 .
n=0 n=1

41 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Short Recap

Power Series, Radius of Convergence, Analyticity, Taylor’s Series

Ordinary Points vs. Singular Points

Power Series Solution, Recursive Formula

Regular Singular Point vs. Irregular Singular Point

Generalized Power Series

Method of Frobenius, Indicial Equation

42 / 43 王奕翔 DE Lecture 9
 Review of Power Series
Solutions about Ordinary Points
Solutions about Singular Points
Summary

Self-Practice Exercises

6-1: 1, 7, 13, 15, 19, 23, 25, 29, 35

6-2: 1, 3, 13, 15, 19, 21, 23

6-3: 1, 3, 5, 9, 11, 13, 17, 25, 27, 29, 33

43 / 43 王奕翔 DE Lecture 9