Subscribe!

• Colins Blog

• Fill In The Gaps
• Software Checklist
• NASA Space Crews
• The Birthday Paradox
• The Trapezium Conundrum
• Revisiting The Ant
• The Ant And The Rubber Band
• Irrationals Exist
• Multiple Choice Probability Puzzle
• Random Eratosthenes
• Wrapping Up Square Dissection
• Dissecting A Square Part 2
• Dissecting A Circle
• Dissecting A Square
• An Oddity In Tennis
• Decision Tree For Tennis
• Decision Trees In Games
• A Matter Of Convention
• Do You Nourish Or Tarnish
• Binary Search Reconsidered
• Two Equals Four
• The Lost Property Office
• The Forgiving User Interface
• Setting Up RSS
• Withdrawing From Hacker News

• Random Writings.
• Colins Blog 2010
• Colins Blog 2009
• Colins Blog 2008
• Colins Blog 2007
• Colins Blog Before 2007

Beyond the Boundary - 2014/05/23

So when we were talking about how to Fill In The Gaps we saw that for the function $f(x)=sin(x)/x$ we didn't have a value for x=0. That's clear from the formula, because when we substitute the value 0 for x we get $0/0,$ and that's undefined.

Actually f is defined on all complex numbers except 0, but for the moment we are concentrating on the reals.

So it brought us to the idea that when we define a function we also need to say what its domain is. We need to say what the values are that we are allowed to feed to the function. In this case we see that f is only defined on the real numbers without 0.

But we can see just by plotting $\frac{sin(x)}{x}$ that it makes sense to define f(0)=1. That fills the gap, the resulting curve flows smoothly, everything seems to be OK.

In fact, we can show that this expression:

• $1-\frac{x^2}{3!}+\frac{x^4}{5!}-\frac{x^6}{7!}+...$

There doesn't seem to be much that's controversial about all that. More, we can make precise what we mean when we say that a particular augmentation of a function "makes sense." There's a process called "Analytic Continuation" that basically means "Extend the existing definition in a way that keeps everything smooth and well-behaved."

Let me show you an example.

Consider the expression:

• $g(x)=1+x+x^2+x^3+x^4+x^5+...$

Think of any value for x - say $\frac{1}{2}.$ Put that in and we get an infinite sum:

If we cut that off early we find that we get the sequence of partial sums:

Technically, let's suppose there's a special location, and every time you give me a permitted error, there comes a point after which all the elements of the sequence are closer. Then if you become more demanding, I just need to go further to find that all subsequent elements satisfy your demands. When that happens, the sequence is said to approach that special location as a "limit."

This sequence gets relentlessly closer to 2, as close as we like. The sequence has 2 as a limit. In this sense we say that the infinite sum converges, and that it makes sense to talk about a value of the infinite sum.

In short, when x is $\frac{1}{2},$ we say that the infinite sum is equal to 2.

You can substitute any value of x between (but not including) -1 and 1, and you find that the infinite sum converges to a value. Because of this we say that the expression:

is well-defined for x on the domain (-1,1) (the round brackets (as opposed to square brackets) showing that we do not include the endpoints) and all is well.

For x=1 and bigger, the infinite sum does not converge. It does not make sense to talk about "the value" of g(x) when $x\le-1$ or $x\ge{1}.$ In particular, when x=2 we get the expression:

and clearly the partial sums don't converge to a limit, they disappear off into the distance.

But now comes the interesting part. Consider the function $h(x)=\frac{1}{1-x}$ which is defined for all x except x=1. When $-1\lt{x}\lt{1}$ we find that the value of g(x) is the same as the value of h(x). In other words, everywhere that g(x) is defined and has a reasonable value, that value is the same as the function h(x).

But h(x) is defined in places where g(x) is not defined. In particular, if we look at x=2 we get:

• $g(x)=1+2+4+8+16+...$
• $h(x)=\frac{1}{1-2}=-1$

We say that the function h(x) extends the function g(x).

Of course, it's easy to take an original function and just randomly specify additional values. That's no real challenge. The difficulty is to do this in a way that is some sense "natural" or "obvious." We saw that in the previous article - Fill In The Gaps - where we added a value for a single point, and it was "natural" for that value to be 1. Anything else felt like it would be contrived.

In a later we make this concept more precise.

<<<< Prev <<<< Fill In The Gaps

:

>>>> Next >>>> Square Root By Long Division ...

---

You should follow me on twitter

Comments

I've decided no longer to include comments directly via the Disqus (or any other) system. Instead, I'd be more than delighted to get emails from people who wish to make comments or engage in discussion. Comments will then be integrated into the page as and when they are appropriate.

If the number of emails/comments gets too large to handle then I might return to a semi-automated system. We'll see.