Subsequences

Introduction: Why subsequences matter 📌

Imagine students is watching a long playlist of song ratings, temperature readings, or stock prices over time. Sometimes the whole list is hard to understand, but a smaller list picked from it can reveal a hidden pattern. In real analysis, that smaller list is called a subsequence. Subsequence ideas are essential because they help us study whether a sequence settles down, jumps around, or has some hidden structure.

In this lesson, students will learn:

what a subsequence is and how to identify one
how subsequences relate to sequences and limits
how to use subsequences in reasoning about convergence
why subsequences are important in proving bigger theorems in real analysis

Subsequences are one of the most useful tools in the study of sequences because they let us zoom in on selected terms without changing their original order. That makes them perfect for spotting behavior in sequences that may not be obvious at first glance. 🌟

What is a subsequence?

A sequence is a list of numbers written in order, such as $\{a_n\}_{n=1}^{\infty}$. A subsequence is formed by choosing terms from the original sequence using a strictly increasing list of indices.

If $\{n_k\}_{k=1}^{\infty}$ is a sequence of natural numbers such that $n_1 < n_2 < n_3 < \cdots$, then the sequence

$\{a_{n_k}\}_{k=1}^{\infty}$

is a subsequence of $\{a_n\}_{n=1}^{\infty}$.

This means the terms appear in the same order as in the original sequence, but some terms may be skipped.

Example 1

Let $a_n = n$. Then the sequence is

$1, 2, 3, 4, 5, \dots$

A subsequence could be the even-indexed terms:

$2, 4, 6, 8, \dots$

This corresponds to $a_{n_k}$ where $n_k = 2k$.

Another subsequence is the terms at prime indices:

$2, 3, 5, 7, 11, \dots$

Here the values are chosen according to the prime-number positions in the original sequence.

Important idea

A subsequence does not rearrange terms. It only selects terms in increasing index order. That is what makes it a valid subsequence.

How subsequences connect to limits

Subsequences are especially important when studying convergence. If a sequence $\{a_n\}$ converges to a limit $L$, then every subsequence also converges to the same limit $L$.

This is a central fact in real analysis:

\text{If } $\lim_{n\to\infty}$ a_n = L, \text{ then } $\lim_{k\to\infty}$ a_{n_k} = L \text{ for every subsequence } \{a_{n_k}\}.

Why is this true? If the original sequence gets arbitrarily close to $L$, then any selected terms from far enough out in the sequence must also be close to $L$. A subsequence cannot escape the behavior of a convergent sequence.

Example 2

Consider

$a_n = \frac{1}{n}.$

We know

$\lim_{n\to\infty} \frac{1}{n} = 0.$

Now take the subsequence $a_{n_k}$ with $n_k = 2k$:

$a_{n_k} = \frac{1}{2k}.$

This subsequence also satisfies

$\lim_{k\to\infty} \frac{1}{2k} = 0.$

If we instead choose $n_k = k^2$, then

$a_{n_k} = \frac{1}{k^2},$

and this also converges to $0$.

So no matter which subsequence students chooses, the limit stays the same as long as the original sequence converges.

Divergent sequences and subsequences

Subsequences become even more powerful when the original sequence does not converge. A divergent sequence may still contain subsequences that converge.

Example 3: alternating sequence

Consider

$a_n = (-1)^n.$

This sequence is

-1, 1, -1, 1, -1, 1, $\dots$

It does not converge because it keeps jumping between $-1$ and $1$.

But it has subsequences that do converge:

the even-indexed subsequence $a_{2k}$ is

$ a_{2k} = 1$

so it converges to $1$

the odd-indexed subsequence $a_{2k-1}$ is

$ a_{2k-1} = -1$

so it converges to $-1$

This example shows that a single sequence can have different subsequences with different limits. That is one reason subsequences are so useful: they reveal hidden patterns inside a sequence that may not converge as a whole. 🔍

Key consequence

If a sequence has two subsequences with different limits, then the original sequence cannot converge. Why? Because if the original sequence had a limit $L$, then every subsequence would have to converge to $L$. Different subsequence limits would contradict that.

How to think about subsequences rigorously

To work with subsequences in real analysis, students should focus on two things:

the indices must increase strictly
the order of the chosen terms must match the original sequence

A subsequence is not just any collection of terms. For example, from $1, 2, 3, 4, 5, \dots$, the list $4, 2, 6 is not a subsequence because the order does not follow the original sequence.

Example 4: deciding whether something is a subsequence

Let $a_n = n^2$, so the sequence is

$1, 4, 9, 16, 25, \dots$

Is $4, 16, 36, 64, \dots$ a subsequence?

Yes. These are the terms $a_2, a_4, a_6, a_8, \dots$, so the index sequence is $n_k = 2k$.

Is $9, 4, 25 a subsequence?

No. Even though all three numbers appear in the original sequence, the order is wrong.

Example 5: subsequence from a formula

Suppose

$a_n = \frac{n}{n+1}.$

If we choose $n_k = k+1$, then the subsequence is

$a_{n_k} = \frac{k+1}{k+2}.$

Since

$\lim_{k\to\infty} \frac{k+1}{k+2} = 1,$

this subsequence converges to $1$. In fact, the whole sequence also converges to $1$.

This shows how subsequences often help confirm a limit or test whether a limit is plausible.

Subsequence reasoning in real analysis proofs

Subsequences are not only examples; they are also tools in proofs. Many important results in real analysis use subsequences to investigate boundedness, convergence, and compactness.

Bolzano–Weierstrass idea

A major theorem says that every bounded sequence in $\mathbb{R}$ has a convergent subsequence. This does not mean the whole sequence converges. It only means that inside a bounded sequence, students can always find a smaller sequence that settles down.

For example, the sequence

$a_n = (-1)^n$

is bounded because all terms lie between $-1$ and $1$. It does not converge, but it has convergent subsequences.

This theorem is important because it tells us that boundedness prevents a sequence from escaping to infinity, and subsequences can expose the structure that remains.

A useful proof idea

If students wants to prove a sequence does not converge, one strategy is to find two subsequences with different limits. For example, for $a_n = (-1)^n$, we can show

$\lim_{k\to\infty} a_{2k} = 1$

and

$\lim_{k\to\infty} a_{2k-1} = -1.$

Since these limits are different, the original sequence cannot converge.

This is a common method in analysis: use subsequences to detect failure of convergence. 🧠

Common mistakes to avoid

When working with subsequences, watch out for these errors:

Changing the order of terms. A subsequence must preserve order.
Using repeated or non-increasing indices. The indices must satisfy

n_1 < n_2 < n_3 < $\cdots.$

Assuming every subsequence of a divergent sequence diverges. That is false. A divergent sequence can still have convergent subsequences.
Assuming one convergent subsequence means the whole sequence converges. That is also false. One convergent subsequence is not enough.

A good habit is to write the index sequence explicitly. That makes it easier to check whether a proposed subsequence is valid.

Conclusion

Subsequences are a foundational idea in real analysis because they let students focus on selected terms of a sequence while keeping the original order. If a sequence converges, every subsequence converges to the same limit. If a sequence does not converge, subsequences may still reveal meaningful limits or show that the original sequence cannot converge.

Subsequences connect directly to the broader study of sequences and limits, especially through convergence tests, proof strategies, and the Bolzano–Weierstrass theorem. They help turn a complicated sequence into something easier to understand and prove. That makes subsequences one of the most powerful tools in the study of real numbers and limits. ✅

Study Notes

A subsequence of $\{a_n\}$ is written as $\{a_{n_k}\}$, where the indices satisfy $n_1 < n_2 < n_3 < \cdots$.
A subsequence keeps the original order of terms and skips none, some, or many terms.
If $\lim_{n\to\infty} a_n = L$, then every subsequence $\{a_{n_k}\}$ also satisfies $\lim_{k\to\infty} a_{n_k} = L$.
If two subsequences of the same sequence have different limits, then the original sequence does not converge.
A bounded sequence in $\mathbb{R}$ has at least one convergent subsequence.
Common examples include $a_n = (-1)^n$, which has subsequences converging to $1$ and $-1$.
To test a proposed subsequence, check both the formula for the terms and the increasing order of the indices.
Subsequence arguments are often used to prove convergence, non-convergence, and boundedness results in real analysis.