How Do Series Converge in Normed Spaces?

  • Thread starter Thread starter Greg Bernhardt
  • Start date Start date
  • Tags Tags
    Convergence Series
Messages
19,773
Reaction score
10,726
Definition/Summary

In what follows, we will work in a normed space (X,\|~\|).
A series is, by definition, two sequences (u_n)_n and (s_n)_n such that s_n=\sum_{k=0}^n{u_k} for every n.

We call the elements u_n the terms of the series. The elements s_n are called the partial sums. We will often denote a series by \sum_{n=0}^{+\infty}{u_n}.

We say that a series \sum_{n=0}^{+\infty}{u_n} converges to a if and only if s_n\rightarrow a. If a series does not converge, then we say that the series diverges.

Equations



Extended explanation

Series in a normed space

For the following, we will work in a normed space (X,\| ~\|)

nth term test
If \sum_{n=0}^{+\infty}{u_n} is a series such that \lim_{n\rightarrow +\infty}{u_n}\neq 0,
then the series diverges.

WARNING: The converse does not hold, i.e. if the limit does equal zero, then the series does not necessarily converge.

HINT: When given a series, always apply this test first.

Linearity of convergence
Let \lambda, \mu\in \mathbb{R}. If \sum_{n=0}^{+\infty}{u_n} converges to u and if \sum_{n=0}^{+\infty}{v_n} converges to v, then the series \sum_{n=0}^{+\infty}{\lambda u_n+\mu v_n} converges to \lambda u+\mu v.

Deletion of finitely many terms
Let p\in \mathbb{N}. Then we have the following equivalence:
\sum_{n=0}^{+\infty}{u_n}~\text{converges iff }~\sum_{n=p}^{+\infty}{u_n}~\text{converges}

Series in a complete normed space

In the following, we will work in a Banach space (= a complete normed space).

Cauchy criterion
A series \sum_{n=0}^{+\infty}{u_n} converges if and only if
\forall \epsilon>0:~\exists n_0:~\forall n>n_0:~\forall p:~\left\|\sum_{k=n}^{n+p}{u_k}\right\|<\epsilon

Absolute convergence
Let \sum_{n=0}^{+\infty}{u_n} be a series. If the series \sum_{n=0}^{+\infty}{\|u_n\|} converges, then the orginal series will converge. Moreover, we have

\left\|\sum_{n=0}^{+\infty}{u_n}\right\|\leq \sum_{n=0}^{+\infty}{\|u_n\|}

NOTATION: A series such as in the above theorem is called absolutely convergent. Absolute convergence is handy because it allows you to transform a series to a series with positive real numbers.

Series with nonnegative real terms

In the following we will always work with series \sum_{n=0}^{+\infty}{u_n} such that all the u_n are real and nonnegative.

Subseries
If \sum_{n=0}^{+\infty}{u_n} is a convergent series and if \sum_{n=0}^{+\infty}{u_{k_n}} is a subseries, then this subseries converges. In particular, we have that

\sum_{n=0}^{+\infty}{u_{k_n}}\leq \sum_{n=0}^{+\infty}{u_n}

Comparison test
Let \sum_{n=0}^{+\infty}{u_n} and \sum_{n=0}^{+\infty}{v_n} be two series such that u_n\leq v_n for all n greater then a certain n_0. Then we have:

1) If \sum_{n=0}^{+\infty}{v_n} converges, then \sum_{n=0}^{+\infty}{u_n} converges.

If \sum_{n=0}^{+\infty}{u_n} diverges, then \sum_{n=0}^{+\infty}{v_n} diverges.

Limit comparison test
1) If \limsup_{n\rightarrow +\infty}{\frac{u_n}{v_n}}<+\infty and if \sum_{n=0}^{+\infty}{v_n} converges, then \sum_{n=0}^{+\infty}{u_n} converges.

2) If \liminf_{n\rightarrow +\infty}{\frac{u_n}{v_n}}>0 and if \sum_{n=0}^{+\infty}{u_n} converges, then \sum_{n=0}^{+\infty}{v_n} converges.

HINT: the limsup and liminf can be replaced by ordinary limits.

Comparison test 2
Let \sum_{n=0}^{+\infty}{u_n} and \sum_{n=0}^{+\infty}{v_n} be series. If there exists an m such that for every n\geq m it holds that \frac{u_{n+1}}{u_n}\leq \frac{v_{n+1}}{v_n}, then

1) If \sum_{n=0}^{+\infty}{v_n} converges, then \sum_{n=0}^{+\infty}{u_n} converges.

1) If \sum_{n=0}^{+\infty}{u_n} diverges, then \sum_{n=0}^{+\infty}{v_n} diverges.

Cauchy condensation test
Let (u_n)_n be a nonincreasing sequence, then

\sum_{n=0}^{+\infty}{u_n}~\text{converges if and only if}~\sum_{n=0}^{+\infty}{2^nu_{2^n}}~\text{converges.}

Cauchy's root test
Let \sum_{n=0}^{+\infty}{u_n} be a series. Then

1) If \limsup_{n\rightarrow +\infty}{\sqrt[n]{u_n}}<1, then \sum_{n=0}^{+\infty}{u_n} converges.

2) If \limsup_{n\rightarrow +\infty}{\sqrt[n]{u_n}}>1, then \sum_{n=0}^{+\infty}{u_n} diverges.

HINT: the limsup can be replaced by ordinary limits.

WARNING: if the limsup equals 1, then the test is inconclusive.

The ratio test of d'Alembert
Let \sum_{n=0}^{+\infty}{u_n} be a series. Then

1) If \limsup_{n\rightarrow +\infty}{\frac{u_{n+1}}{u_n}}<1, then \sum_{n=0}^{+\infty}{u_n} converges.

2) If \liminf_{n\rightarrow +\infty}{\frac{u_{n+1}}{u_n}}>1, then \sum_{n=0}^{+\infty}{u_n} diverges.

HINT: the limsup and liminf can be replaced by ordinary limits.

WARNING: if the limits equal 1, then the test is inconclusive.

The integral test
Let f:[0,+\infty[\rightarrow\mathbb{R}^+ be a nonincreasing function. Then

\sum_{n=0}^{+\infty}{f(n)}~\text{converges if and only if}~\int_1^{+\infty}{f(x)dx}<+\infty

ADDENDUM: If f:[0,+\infty[\rightarrow\mathbb{R}^+ is a nonincreasing function, then for every n\in \mathbb{N} holds

\sum_{k=1}^n{f(k)}\leq \int_0^n{f(x)dx}\leq \sum_{k=0}^{n-1}{f(k)}


Series in \mathbb{R} and \mathbb{C}

The criterion of Dirichlet
Let \sum_{n=0}^{+\infty}{a_n} be a (real or complex) series such that it's sequence of partial sums is bounded. Let (v_n)_n be a nonincreasing sequence of real numbers which converges to 0. Then the sequence \sum_{n=0}^{+\infty}{v_na_n} converges.

The criterion of Abel
Let \sum_{n=0}^{+\infty}{a_n} be a (real or complex) convergent series . Let (v_n)_n be a bounded sequence of real numbers which is either nondecreasing or nonincreasing. Then the sequence \sum_{n=0}^{+\infty}{v_na_n} converges.

The criterion of Leibniz
Let (u_n)_n be a nonincreasing sequence of real numbers which converges to 0. Then the series \sum_{n=0}^{+\infty}{(-1)^nu_n} converges.

ADDENDUM: Denote (s_n)_n the partial sums of the series \sum_{n=0}^{+\infty}{(-1)^nu_n} and denote s the limit of the series. Then the sequence (s_{2n})_n is nonincreasing and (s_{2n+1})_n is nondecreasing. Moreover, we have that |s-s_n|\leq u_n.

Some special series
Geometric series
Let x be an arbitrary real or complex number. Then

\sum_{n=0}^{+\infty}{x^n}~\text{converges if and only if}~|x|<1.

Moreover, if the series converges, then \sum_{n=0}^{+\infty}{x^n}=\frac{1}{1-x}


p-series
Let p be a real number. Then

\sum_{n=0}^{+\infty}{\frac{1}{n^p}}~\text{converges if and only if}~p>1.

* This entry is from our old Library feature. If you know who wrote it, please let us know so we can attribute a writer. Thanks!
 
Mathematics news on Phys.org
Greg Bernhardt said:
* This entry is from our old Library feature. If you know who wrote it, please let us know so we can attribute a writer. Thanks!
Hey, @Greg Bernhardt ,
For these old Library features, you may want to place the above message at the beginning of each post/thread.

SammyS
 
SammyS said:
For these old Library features, you may want to place the above message at the beginning of each post/thread.
You'd think it would be that easy, but then it screws up the meta description because it takes from the beginning of the first post :frown:
 
In medias res is better. Nobody wants to read any technical preliminaries, so the comment at the end makes more sense.
 
Greg Bernhardt said:
You'd think it would be that easy, but then it screws up the meta description because it takes from the beginning of the first post :frown:
Yup. That makes sense.

I see that you are not getting much info. regarding the authorship of these, but I'm learning a lot from @fresh_42's comments on these old Library threads and so many others.

Keep it going fresh! You too, Mark44 & others.
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
I'm interested to know whether the equation $$1 = 2 - \frac{1}{2 - \frac{1}{2 - \cdots}}$$ is true or not. It can be shown easily that if the continued fraction converges, it cannot converge to anything else than 1. It seems that if the continued fraction converges, the convergence is very slow. The apparent slowness of the convergence makes it difficult to estimate the presence of true convergence numerically. At the moment I don't know whether this converges or not.
Back
Top