Fundamentals of Series: Table II: Examples of Series

Which Appear in Calculus

From the seven unpublished manuscripts of H. W. Gould
Edited and Compiled by Jocelyn Quaintance
May 3, 2010

1 The Binomial Theorem

Remark 1.1 In this table, unless otherwise specified, n and r are nonnegative intergers, and x
and z are arbitrary real or complex numbers. We also assume that for any real number x, [x] is
the greatest integer in x.

1.1 Binomial Theorem

1.1.1 Basic Form with Integer Power
n  
n
X n n−k k
(x + a) = x a , where a is an arbitrary real or complex number (1.1)
k=0
k

1.1.2 Newton’s Binomial Theorem

∞  
z
X z k
(1 + x) = x , where z is a real or complex number and |x| < 1 (1.2)
k=0
k

1.1.3 Applications of Newton’s Binomial Theorem

∞ 
2k xk

X 1
=√ , |x| < 1 (1.3)
k=0
k 22k 1−x

∞ 
√

X 2k 1
3k
= 2 (1.4)
k=0
k 2

1
 ∞   √
X 2k 1 2
(−1)k 2k
= (1.5)
k=0
k 2 2

∞ 
xk √

X 2k
− = 1 − x, |x| < 1 (1.6)
k=0
k 22k (2k − 1)

∞  
X 2k 1
=0 (1.7)
k=0
k 22k (2k − 1)

∞ 
(−1)k √

X 2k
= − 2 (1.8)
k=0
k 22k (2k − 1)

1.2 Companion Binomial Theorem

Remark 1.2 In Section 1.2, we assume p is a nonnegative integer. We also assume a and b are
arbitrary real or complex numbers.

Companion Binomial Theorem

∞   ∞  
X n+p n X n n−p 1
x = x = p+1
, |x| < 1 (1.9)
n=0
n n=p
p (1 − x)

1.2.1 Applications of Companion Binomial Theorem

∞   k
1 1 X k k+p b b
= p+1 (−1) , | |<1 (1.10)
(a + b)p+1 a k=0
k a k a

∞ 
1 X k + p bk

1 b
= p+1 , | |<1 (1.11)
(a − b)p+1 a k=0
k ak a

∞  
1 X
k k+p
= (−1) xmk , |x| < 1, m ∈ < (1.12)
(1 + xm )p+1 k=0
k

2
 ∞  
1 X k + p mk
= x , |x| < 1, m ∈ Re (1.13)
(1 − xm )p+1 k=0
k

∞  
1 X
k k+p
= (−1) x−m(k+p+1) , |x| > 1, m ∈ < (1.14)
(xm + 1)p+1 k=0
k

∞  
1 X k + p −m(k+p+1)
= x , |x| > 1, m ∈ < (1.15)
(xm − 1)p+1 k=0
k

1.2.2 Finite Version of Companion Binomial Theorem

n  
X n+k 1
k
= 2n (1.16)
k=0
k 2

Variation of Finite Companion Binomial Theorem

n  
X 2n − k k
2 = 22n (1.17)
k=0
n

Application of Finite Companion Binomial Theorem

∞  
X 2n + k 1
k
= 22n (1.18)
k=1
n 2

3
1.3 Binomial Theorem with Complex Exponents
Remark 1.3 The material in Section 1.3 is found in T. J. I’a. Bromwich’s Introduction to the
Theory of Infinite Series, Second Edition, 1949, Chapter 9, Article 96.

Let α, β ∈ <. Let i2 ≡ −1. Then,

∞  
α+βi
X α + βi
(1 + x) = xk , (1.19)
k=0
k

where

a. The series is absolutely convergent for |x| < 1.

b. If α > 0, the series converges absolutely on the circle |x| = 1. Hence, the series is uniformly
convergent within and on the circle |x| = 1.

c. If −1 < α ≤ 0, the series converges on the circle |x| = 1 except at x = −1.

d. If α ≤ −1, the series diverges everywhere on the circle |x| = 1.

1.4 Applications of the Binomial Theorem

1.4.1 Derivatives of the Binomial Series
n     
X n k k r n−r
n
x = x (1 + x) (1.20)
k=0
k r r

n     
X n k n−r n
=2 (1.21)
k=0
k r r

n     
X n k k r n−r n
2 =2 3 (1.22)
k=0
k r r

n     
X
k
n k k r r n−r
n
(−1) x = (−1) x (1 − x) (1.23)
k=0
k r r

n  
X n k
(−1) k2k = 2n(−1)n (1.24)
k=0
k

4
 1
1.4.2 Expansions of (1 − x)− 2

∞  −1 
X
k 1
(−1) 2 xk = √ , |x| < 1 (1.25)
k=0
k 1−x

∞ 
xk

X 2k 1 x
= , |x| < 1 or | |<1 (1.26)
k=0
k (1 + x)2k+1 1−x 1+x

∞    k  x r 2r
X 2k k x − 2r+1
= (1 − x) 2 , |x| < 1 (1.27)
k=0
k r 22k 4 r

∞    √  
X 2k k 1 2 2r
3k
= 2r (1.28)
k=0
k r 2 2 r

∞   √
X 2k k 2
3k
= (1.29)
k=0
k 2 2

∞  
X 2k k k x
2k
x = 3 , |x| < 1 (1.30)
k=0
k 2 2(1 − x) 2

∞  
X 2k 2k + 1 1
xk = 3 , |x| < 1 (1.31)
k=0
k 22k (1 − x) 2

∞ 
√

X 2k 2k + 1
= 8 (1.32)
k=0
k 23k
Bruckman’s Formula Version 1
n  −1  −1 
X
2 2
1 −1
= −1 (1.33)
k n − k (2k + 1)(2n − 2k + 1)


k=0 2(n + 1)2 2
n+1

Bruckman’s Formula Version 2

n 
24n+1
 
X 2k 2n − 2k 1
= (1.34)
(n + 1)2 2n+2


k=0
k n−k (2k + 1)(2n − 2k + 1) n+1
24n
= 2n


(n + 1)(2n + 1) n

5
 1
1.4.3 Expansions of (1 − x) 2

∞ 
xk

X 2k 1

k
= −(1 − x) 2 , |x| < 1 (1.35)
k=0
k 4 (2k − 1)

∞  1
2k + 1 z k+1

X 1 − (1 − 4z) 2 1
= , |z| < (1.36)
k=0
k 2k + 1 2 4

Pn n xk


1.4.4 Evaluation of k=0 k k+1
n  
X n xk (x + 1)n+1 − 1
= , x 6= 0 (1.37)
k=0
k k+1 (n + 1)x

n  
X n 1 2n+1 − 1
= (1.38)
k=0
k k+1 n+1

n  
X n k1 1
(−1) = (1.39)
k=0
k k+1 n+1

2n   k
X 2n 2
k 1
(−1) = (1.40)
k=0
k k+1 2n + 1

2n
2n + 1 2k 22n
X  
k
(−1) = (1.41)
k=0
k k+1 n+1

2n  k
22n

X
k 2n 2 1
(−1) = − , n≥1 (1.42)
k=1
k−1 k+1 n + 1 2n + 1

n
[2]   2k
X n x (x + 1)n+1 − (1 − x)n+1
= , x 6= 0 (1.43)
k=0
2k 2k + 1 2(n + 1)x

6
 n
[2] 
2n

X n 1
= (1.44)
k=0
2k 2k + 1 n+1

[ n−1
2
]  2k
X n x (x + 1)n+1 + (1 − x)n+1 − 2
= , x 6= 0 (1.45)
k=0
2k + 1 k + 1 (n + 1)x2

[ n−1
2
]
2n+1 − 2

X n 1
= (1.46)
k=0
2k + 1 k + 1 n+1

n   n+r
Pr−1 n+r

k

X n k! n! (x + 1) − k=0 x
xk = k
, x 6= 0, r ≥ 1 (1.47)
k=0
k (k + r)! (n + r)!xr

n   k+r
(x + 1)n+r − r−1 n+r
P
i
X n x x
k+r

= n+r

i=0 i
, r≥1 (1.48)
k=0
k k n

n   r−1
X nk k! r+1
X 1
(−1) = (−1) n! (−1)k , r≥1 (1.49)
k=0
k (k + r)! k=0
(n + r − k)!k!

n r−1  !
xn+r
  
k n 1 k + r
X X
(−1) k+r

(1 − x)k+r − (−1)α xα = (−1)r n+r
, r ≥ 1 (1.50)
k=0
k k α=0
α n

1.4.5 Expansions of (t − a)n−1 (t − (a + nx))

Remark 1.4 In the identities related to the expansion of (t − a)n−1 (t − (a + nx)) , we assume,
unless otherwise specified, that x, t, and a are nonzero real or complex numbers.
n  
X
k
n
(−1) tn−k ak−1 (a + kx) = (t − a)n−1 (t − (a + nx)), n≥1 (1.51)
k=0
k

n  
X
kn n−k k−1
lim (−1) t a (a + kx) = tn − nxtn−1 , n≥1 (1.52)
a→0
k=0
k

7
 
n   0,
 n 6= 0, 1
k n
X
n−1
a (−1) (a + kx) = −x, n=1 (1.53)
k=0
k 
1, n=0


n  
X n
(−1) (f (x))k (a + kx) = (1.54)
k=0
k
(1 − f (x))n−1 (a − (a + nx)f (x)) , n≥1

n  
X n
(−1) (f (x))k (a − kx) = (1.55)
k=0
k
(1 − f (x))n−1 (a − (a − nx)f (x)) , n ≥ 1

n    n−1 
k n a + kx 1
X a 
(−1) = 1− a− +x , n≥1 (1.56)
k=0
k nk n n

n  
X
k n a + kx a−x
lim (−1) k
= (1.57)
n→∞
k=0
k n e

n    n−1 
X n a + kx 1 a 
= 1+ a+ +x , n≥1 (1.58)
k=0
k nk n n

n  
X n a + kx
lim = (a + x)e (1.59)
n→∞
k=0
k nk

n  
X n a + kx
k a − nx
(−1) = (1.60)
k=0
k 2k 2n

n    n−1  
k n a + kx 1 a + nx
X
(−1) nk
= 1− n a− n
, n≥1 (1.61)
k=0
k 2 2 2

8
 n  
X
kn a + kx
lim (−1) =a (1.62)
n→∞
k=0
k 2nk

n    n−1
k n a + kx 1
X
(−1) = 1− (a − 1), n≥1 (1.63)
k=0
k (a + nx)k a + nx

n    n−1  
X
k n a + kx 1 a + nx
(−1) = 1− a− , n≥1 (1.64)
k=0
k (a − nx)k a − nx a − nx

n  
X n
k−1
(−1) k (f (x))k = nf (x) (1 − f (x))n−1 (1.65)
k=0
k

n  
k−1 n
X
(−1) k 2 (f (x))k−1 = n (1 − f (x))n−1 − n(n − 1)f (x) (1 − f (x))n−2 (1.66)
k=0
k

n    2 
X n 2 n n +n
k =2 (1.67)
k=0
k 4

n  
X n 2 k−1
k 2 = 3n−2 n(2n + 1) (1.68)
k=0
k

n  
X n
ak k = na(1 + a)n−1 (1.69)
k=0
k

n  
X n
k = n2n−1 (1.70)
k=0
k

n  
X n
k(k − 1)xk−2 = n(n − 1)(1 + x)n−2 (1.71)
k=0
k

n  
X n
k(k − 1) = n(n − 1)2n−2 (1.72)
k=0
k

9
1.4.6 Number Theoretic Result Due to Euler

n
X
Let f (x) = ai x i . Then, f (x)|f (x + f (x)). (1.73)
i=0

1.5 Four Versions of the Multinomial Theorem

Remark 1.5 In Section 1.5, we will assume α is a nonnegative integer. We also assume that ji is
a nonnegative integer.

n
!α
X X α!
ai = aj00 aj11 ...ajnn (1.74)
i=0 ∀jsuch that
j0 !j1 !j2 !...jn !
P n
i=0 ji =α

n
!α
X X α!
ai = aj1 aj2 ...ajnn (1.75)
i=1 ∀jsuch that
j1 !j2 !...jn ! 1 2
P n
i=1 ji =α

n
!α NX
<∞
X
i
X α!
ai x = xk aj00 aj11 ...ajnn (1.76)
i=0 k=0 ∀jsuchPthat
j0 !j1 !j2 !...jn !
Pn n
i=0 ji =α, i=1 iji =k

n
!α NX
<∞
X
i
ai x = xk ∗
i=0 k=0
 
X α α−γ γ!
a0 aj1 aj2 ...ajnn (1.77)
∀jsuch that P
γ j1 !j2 !...jn ! 1 2
Pn n
j0 +γ=α, i=1 ji =γ, i=1 iji =k

10
2 The Geometric Series
Remark 2.1 In this chapter, we will assume, unless otherwise specified, that a is a nonnegative
integer and x is an arbitrary nonzero real or complex number.

2.1 The Basic Geometric Series

2.1.1 Finite Geometric Series
n
X xn−a+1 − 1
xk = xa , x 6= 1 (2.1)
k=a
x−1

n
X 1 1 xn−a+1 − 1
= , x 6= 1 (2.2)
k=a
xk xn x − 1

2.1.2 Infinite Geometric Series

∞
X xa
xk = , |x| < 1 (2.3)
k=a
1−x

∞
X x1−a
x−k = , |x| > 1 (2.4)
k=a
x−1

2.2 Derivatives of Geometric Series

n
X nxn+1 − (n + 1)xn + 1
kxk−1 = , x 6= 1 (2.5)
k=0
(x − 1)2

n n
X n2 + n X
lim kxk = = k (2.6)
x→1
k=0
2 k=0

n
X n2 xn+3 − (2n2 + 2n − 1)xn+2 + (n + 1)2 xn+1 − x2 − x
k 2 xk = , x 6= 1 (2.7)
k=0
(x − 1)3

n
X
k(k − 1)2k = (n2 − 3n + 4)2n+1 − 23 (2.8)
k=0

11
 n
X
k 2 2k = (n2 − 2n + 3)2n+1 − 6 (2.9)
k=0

n
X (n2 − n + 1)3n+1 − 3
k 2 3k = (2.10)
k=0
2

2.3 Integrals of Geometric Series

∞
X xk+1
ln(1 + x) = (−1)k , |x| < 1 (2.11)
k=0
k+1

∞
X 1
ln 2 = (−1)k−1 (2.12)
k=1
k

2.4 Applications of Geometric Series

Remark 2.2 In the following two identities, let u and v be arbitrary nonzero real or complex
numbers such that uv 6= 1.

[ n−1
2
] [ n−2
2
] n+1 n
X
k
X (uv)[ 2 ] − 1
k (uv)[ 2 ] − 1
(uv) + u (uv) = +u , n≥1 (2.13)
k=0 k=0
uv − 1 uv − 1

Remark 2.3 The following identity can be done as a formal calculation over the ring of power
series. Otherwise, the reader may assume that appropriate condition hold so that the left sum is
absolutely convergent.
∞ ∞ k
X X 1 X
f (k) = 2i f (i) (2.14)
k=0 k=0
2k+1 i=0

∞ k
X x2 1
k+1 = , |x| > 1 (2.15)
k=0
1−x 2 1−x

∞ k
X x2 x
k+1 = , |x| < 1 (2.16)
k=0
1−x 2 1−x

12
3 Bernoulli-Type Series and the Riemann Zeta Function

Remark 3.1 In this chapter, we will assume p and a are, unless otherwise specified, nonnegative
integers.

Pn
3.1 Evaluation of k=1 kp
3.1.1 Reduction Formula

n
X n
X n−1 X
X r
p p−1
k =n k − k p−1 , p ≥ 1, n ≥ 1 (3.1)
k=a k=a r=a k=a

3.1.2 Iteration Formulas

Pn
Remark 3.2 In this subsection, we let r be a positive integer, and define (r) f (k) to be the
following r−fold sum:
n
X n
X kr
X k3 X
X k2
f (k) = ... f (k) >
(r) kr =1 kr−1 =1 k2 =1 k=1

n  
X n+r−1
1= (3.2)
r
(r)

n  
X n+r
k= (3.3)
r+1
(r)

X n(n + 1)
k= (3.4)
2
(1)

n
X (2n + r) (n + r)!
k2 = (3.5)
(r + 2)! (n − 1)!
(r)

n
X n(n + 1)(2n + 1)
k2 = (3.6)
3!
(1)

13
 n
X 6n2 + r(6n + r − 1) (n + r)!
k3 = (3.7)
(r + 3)! (n − 1)!
(r)

n
X n2 (n + 1)2
k3 = (3.8)
4
(1)

n
X (12n2 + 12rn − r(5 − r))(2n + r) (n + r)!
k4 = (3.9)
(r + 4)! (n − 1)!
(r)

n
X n(n + 1)(2n + 1)(3n2 + 3n − 1)
k4 = (3.10)
30
(1)

n
X n2 (n + 1)2 (2n2 + 2n − 1)
k5 = (3.11)
12
(1)

n
X n(n + 1)(2n + 1)(3n4 + 6n3 − 3n + 1)
k6 = (3.12)
42
(1)

3.1.3 Euler’s Expansion with Bernoulli Numbers

Remark 3.3 In this subsection, we let Bk denote the k th Bernoulli number. The exponential gen-
erating function of (Bk )∞ x
n=0 is ex −1 . For values of the Bernoulli number sequence, the reader is
referred to the Online Encyclopedia of Integer Sequences (OEIS).
n−1 p
X
p
X p!np−k+1
k = Bk , n≥1 (3.13)
k=0 k=0
(p − k + 1)!k!

3.1.4 G. P. Miller’s Determinant Expansion

n
X det M
kp = , (3.14)
k=0
(p + 1)!
where M is the (p + 1) × (p + 1) matrix whose entry ai,j is determined by

p+2−i
(n + 1)
 − (n + 1), j=1
p+2−i


ai,j = j+1−i
, j ≥ i and j 6= 1 (3.15)

0, j < i and j 6= 1


14
 Pn
3.2 Evaluation of k=0 k p xk
Remark 3.4 The reader should compare the formulas in this subsection with those in Section 2.2.

3.2.1 Differential Reduction Formula

n n
X
p+1 k d X p k
k x =x k x (3.16)
k=0
dx k=0

3.2.2 Applications of Differential Reduction Formula

n
X N
k 3 xk = , where for x 6= 1, (3.17)
k=0
(x − 1)4
N = n3 xn+4 − (3n3 + 3n2 − 3n + 1)xn+3 + (3n3 + 6n2 − 4)xn+2 − (n + 1)3 xn+1 + x3 + 4x2 + x

n−1  k−1
X n
k = (n − 1)2 , n≥1 (3.18)
k=1
n − 1

n+1 k−1
n
!
Y
1+ n

k−2 = n2 , n≥2 (3.19)
k=2 k − k+1
k

n n  
X k k−1
X k n+1
k−1
(n + 1) = k−1
= n2 n≥1 (3.20)
k=1
n k=1
n k+1

Pn n
p
3.3 Evaluation of k=0 k k
n  
X n
= 2n (3.21)
r r=0
n  
X n
r = n2n−1 (3.22)
r=0
r

n  
X n 2
r = 2n−2 n(n + 1) (3.23)
r=0
r

15
 n  
X n 3
r = 2n−3 n2 (n + 3) (3.24)
r=0
r

n  
X n 4
r = 2n−4 n(n + 1)(n2 + 5n − 2) (3.25)
r=0
r

3.3.1 Reduction Formula

n−1   p n   
X n − 1 p XX k p n p−k+1
n r = (−1) r , n≥1 (3.26)
r=0
r k=0 r=0
k r

P∞ 1
3.4 Riemann Zeta Function: ζ(p) = k=1 kp
3.4.1 Convolution Identity
Remark 3.5 The following identity is found in “A New Method of Evaluating ζ(2n)”, by G.T.
Williams, Amer. Math. Monthly, January 1953, Vol. 60, No. 1, pp. 19-25.

n−1 ∞ k
X X 1 X1
ζ(k)ζ(n − k + 1) = (n + 2)ζ(n + 1) − 2 , n≥3 (3.27)
k=2 k=1
k n k=1 j

Extension of Convolution Identity

∞ k
X 1 X1
4ζ(3) − 2 2
=0 (3.28)
k=1
k j=1
j

3.4.2 Connections with Bernoulli Numbers

Remark 3.6 Recall that Bn is the nth Bernoulli number. See Remark 3.3.

(2n)!
B2n = (−1)n−1 ζ(2n) (3.29)
22n−1 π 2n

n−1  
X 1
ζ(2k)ζ(2n − 2k) = n+ ζ(2n), n≥2 (3.30)
k=1
2

16
4 Finite Harmonic Series
4.1 Special Case of nth Difference Inversion Formula
n  
X
k
n 1 1
(−1) = (4.1)
k=0
k k+1 n+1

n n  
X 1 X
k+1 n 1
= (−1) , n≥1 (4.2)
j=1
j k=1
k k

n  Xk
k−1 n 1 1
X
(−1) = , n≥1 (4.3)
k=1
k j=1 j n

4.2 Even and Odd Finite Harmonic Series

2n 2n n
X 1 2n + 1 X 1 X 1
= = (2n + 1) , n≥1 (4.4)
k=1
k 2 k=1 k(2n − k + 1) k=1
k(2n − k + 1)

2n+1 n
X 1 X 1 1
= (2n + 2) + (4.5)
k=1
k k=1
k(2n − k + 2) n + 1

4.3 Harmonic Series as Limit of Binomial Coefficient

Remark 4.1 In the following indentity, we let r be a positive integer.

n n+r


X 1 r
−1
= lim n+r

(4.6)
k=1
k r→0 r r

17