Proving Cauchy Sequence Subsequences

[clueles]

I need help on trying to prove

that every subsequence of a cauchy sequence is a cauchy sequence

 

[Jimmy Snyder]

If $\{x_{n_{k}}\}$ is a subsequence of {$$x_{n}$$}, then for all $$k$$, $$n_{k} \ge k$$. So the '$$N$$' that works for {$$x_{n}$$} also works for {$$x_{n_{k}}$$}

 

[clueles]

is that how you would write a formal proof or would you the for any epsilon>0 there exists an n in i such that abs(sm-sn)<epsilon

 

[Jimmy Snyder]

The answer to that depends on the audience. What is your situation?

 

[clueles]

college level for a real analysis class

 

[Jimmy Snyder]

Please note the small change I made in the proof. At the undergraduate level, probably the full $$\epsilon$$ treatment would be best. Are you up to it?

 

[clueles]

Yes we are up to that part but I'm not sure how to even start it. Do we need to say for all $$\epsilon$$> there exists a n,m /geqN such that abs(Snk-Smk)<E?

 

[Galileo]

I'd start with the definition of a subsequence.
Also, always write down the hypothesis (what you know) and what you need to prove:

Let $\{x_{n_k}\}$ be a subsequence of $\{x_n\}$.

You are given that for every $\epsilon>0$, there exists an integer N>0, such that:

[tex]m,n > N \Rightarrow |x_n-x_m|<\epsilon[/itex]<br /> <br /> You need to prove that for every $\epsilon>0$ there exists an integer N>0, such that:<br /> <br /> $$m_k,n_k > N \Rightarrow |x_{n_k}-x_{m_k}|<\epsilon$$<br /> <br /> All I've done is written down the problem in different terms, but the answer is practically staring you in the face (hopefully).[/tex]

 

[Jimmy Snyder]

$\{x_{n_k}\}$

I'm just starting to use tex and I only know what I see in other people's examples. Your stuff is nice and I learned a lot from it. Do you know of a Windows utility that will take a tex source file and create a pdf file from it?

 

[Galileo]

I'm just starting to use tex and I only know what I see in other people's examples. Your stuff is nice and I learned a lot from it. Do you know of a Windows utility that will take a tex source file and create a pdf file from it?

Yes, WinEdt will do that for you. I think you have to register it or something. I use it often to create pdf's.
The only advantage of not registering is annoying popup reminders every 15 seconds or so. I actually use a different editor and use only WinEdt for creating the pdf's.

 

[Jimmy Snyder]

Thanks, I will look into winedt. In the meantime, I found a fairly large package called miktex which has a utility with the unlikely name of pdftex that turns tex into pdf.

 

[Data]

TeXnic Center is the IDE that I use in general. It has full TeX-to-pdf capability and is built to work with Miktex as well (it requires a TeX install to work, of course). Free too~

It's a little hard to find these days though (although it is quite popular).

If you want, I can probably find somewhere to host it for a few days. It's only about 5mb.

 

[matt grime]

You don't, as far as I can recall (linux for the last 3 years), need any of those IDE's or such. To use latex on MS you need Miktex installed. That comes with the program pdflatex, I think, that creates the pdf for you. It means learning to love the command line though. just type

pdflatex foo.tex

from inside the directory where foo.tex is stored and it outputs a pdf file for you. The installation also comes with dvips which creates a post script file from the pdf. There are other format generators too, including latex2html, though I've no idea if anyone acutally uses this anymore.