MA106 Cauchy Condensation Test
MA106 Cauchy Condensation Test
Luke Peachey
Recall, when showing that the harmonic series diverges (Week 7, Assignment 6), we
grouped the terms as follows:
1 1 1 1 1 1 1
(1) + ( ) + ( + ) + ( + + + ) + ··· ≥
2 3 4 5 6 7 8
1 1 1 1 1 1 1
(1) + ( ) + ( + ) + ( + + + ) + ···
2 4 4 8 8 8 8
We generalise this method of grouping terms to create a convergence test.
Let Ak := {n ∈ N : 2k ≤ n < 2k+1 } ∀k ∈ N0 .
This gives a partition of the naturals N = k∈N0 Ak , with |Ak | = 2k . That is,
F
Notice that this inequality relied only on the fact that ( n1 ) is a decreasing sequence. This
leads to the following generalisation.
Theorem 0.1 (Cauchy Condensation Test). Let (an ) be a decreasing sequence of non-
negative numbers. Then,
∞
X ∞
X
an converges ⇐⇒ 2k a2k converges (4)
n=1 k=0
P∞ k
We call the series k=0 2 a2k the condensed series.
1
Proof. Since (an ) is decreasing, we have the similar inequalities to (1),(2) and (3). That is,
a2k+1 ≤ an ≤ a2k ∀k ∈ N0 , ∀n ∈ Ak .
X X X
2k a2k+1 = a2k+1 ≤ an ≤ a2k = 2k a2k ∀k ∈ N0 .
n∈Ak n∈Ak n∈Ak
N
X N X
X N X
X N X
X N
X
2k a2k+1 = a2k+1 ≤ an ≤ a2k = 2k a2k
k=0 k=0 n∈Ak k=0 n∈Ak k=0 n∈Ak k=0
Since an ≥ 0, our partial sums are increasing. Suppose the condensed series converges.
Then, by the RHS of (5), our partial sums are also bounded, and by completeness, our series
converges. Conversely, suppose the condensed series doesnt converge. Since the partial sums
of the condensed series are increasing, it must diverge to infinity. By the shift rule, the LHS
of (5) goes to infinity also. Therefore, by the one sided sandwich rule, our series diverges.
Example 0.2. Consider the harmonic series as before. We saw that the condensed series is
∞
X ∞
X
k
2 a2k = 1
k=0 k=0
Since this diverges, the Cauchy Condensation Test tells us the harmonic series diverges.
(
∞ 0 :n=1
P 1
Example 0.3. Consider the series n log(n) so that an = 1
n=2 n log(n) :n≥2
The condensed series is then
∞ ∞ ∞
X
k
X 2k 1 X1
2 a2k = =
2k log(2k ) log(2) k
k=1 k=1 k=1
∞
P 1
Since this diverges, by the Cauchy Condensation Test, n log(n) diverges.
n=2
(
∞ 0 :n=1
P 1
Example 0.4. Consider instead the series n log(n)2
so that an = 1
n=2 n log(n)2
:n≥2
The condensed series is then
∞ ∞ ∞
X X 2k 1 X 1
2k a2k = k k 2
= 2
2 log(2 ) log(2) k2
k=1 k=1 k=1
∞
P 1
Since this converges, by the Cauchy Condensation Test, n log(n)2
converges.
n=2