5

I wanted to find the approximate average of all real points in $(x)^{x}$ from $[-4,-2]$. This means I am ignoring all real inputs that give a complex output and need average to be a real number.

To first solve this I found the following defined sets of $x^x$ when $x<0$. $$x=-\frac{2m}{2k+1}|m,k\in\mathbb{Z}$$

$$x=-\frac{2m+1}{2k+1}|m,k\in\mathbb{Z}$$

Each of the sets coincides with another equation so I got the following peice-wise definition. $$x^x=\begin{cases} (-x)^x & x=\left\{ -{2m\over 2k+1}\ |\ m, k \in \Bbb Z\right\}\\ -\left(-x\right)^{x} & x=\left\{ -{2m+1\over 2k+1}\ |\ m, k \in \Bbb Z\right\}\ \\ \text{undefined} & x\neq\left\{ -{2m\over 2k+1}\bigcup-{2m+1\over 2k+1}\ |\ m, k \in \Bbb Z\right\} \end{cases} $$

Then I took the limit definition of an integral along with $\frac{1}{b-a}$ to determine the average. $$\frac{1}{b-a}\lim_{n\to\infty}\sum_{i=1}^{n}{f\left(a+\left(\frac{b-a}{n}\right)i\right)}\left(\frac{b-a}{n}\right)=\frac{1}{b-a}\int_{a}^{b}f(x)$$ where $f(x)=x^x$

(Note that it is possible to skip $x$-values that given an undefined output in an interval. An example is $\{-12/3,-11/3,-10/3,-9/3,-8/3,-7/3,-6/3\}$)

When $n\in\left\{\left.{4n+2}\right| n \in \mathbb{Z} \right\}$ there are numbers whose outputs coincide with $(-x)^{x}$ and numbers whose outputs coincide with $-(-x)^{x}$. So I took the probabilty of having numbers whose ouputs coincide with ($(-x)^x$ vs a numbers whose outputs coincide with $-(-x)^{x}$ in $[-4,-2]$. I got $\frac{1}{2}$ for both of them.

Thus I found when $n \in \left\{\left.{4n+2}\right| n \in \mathbb{Z} \right\}$ as $n\to\infty$

$$\frac{1}{(-2)-(-4)}\lim_{n\to\infty}\sum_{i=1}^{n}{f\left(-4+\left(\frac{2}{n}\right)i\right)}\left(\frac{2}{n}\right)=\frac{1}{2}\left(\frac{1}{2}\int_{-4}^{-2}{(-x)}^{x}+\frac{1}{2}\int_{-4}^{-2}-{\left(-x\right)}^{x}\right)=0$$

But when $n$ is odd, all numbers in the interval have outputs coinciding only with $(-x)^{x}$. So when $n$ is odd integer and $n\to\infty$

$$\frac{1}{-2+4}\lim_{n\to\infty}\sum_{i=1}^{n}{f\left(-4+\left(\frac{2}{n}\right)i\right)}\left(\frac{2}{n}\right)=\frac{1}{2}\int_{-4}^{-2}{\left(-x\right)}^{x}\approx.062152$$

I have two different numbers depending on which $n$value I choose. So does this mean the average does not exist? If not how could one find the average?

What other methods can be used to find a real number average?


EDIT Analyzing this I believe that one should choose specific $n's$ despite a limit having to accept all $n's$

I presume that an interval of reimmmen sum interval should be as dense as possible and represent all defined numbers. This leads me to believe the average is 0.

Since this goes against the definition of a limit, I think the definition of a reimmen sum should be slightly altered for dense defined inputs with undefined outputs.

For example I believe intervals with numerator increase of $1$ in $\{-12/3,-11/3,-10/3,-9/3,-8/3,-7/3,-6/3\}$ can would give an accurate average compared to an numerator increase of $2$ in $\{-12/3,-10/3,-8/3,-6/3\}$.

When there is numerator increase of $1$ I end up with zero. Since it represents inputs whose output coincide with both $(-x)^x$ and $-(-x)^{x}$ the average should be zero.

Perhaps my reasoning is correct but I would like to find out if this is the case.

Arbuja
  • 1
  • 1
    According to http://math.stackexchange.com/questions/398224/what-is-the-domain-of-xx-when-x0 a domain does exist. – Arbuja Feb 04 '16 at 19:22
  • Hopefully this clarifies everything :). – Arbuja Feb 04 '16 at 21:25
  • 1
    Just out of curiosity, why do you want to find this? – YoTengoUnLCD Feb 04 '16 at 21:30
  • 2
    @YoTengoUnLCD
    I became interested in everything related to $x^x$ and its negative domain from its graph, to its bizarre derivative and now with its average. It leads me into stranger areas.
    – Arbuja Feb 04 '16 at 22:51
  • 1
    @Arbuja Why is its derivative bizarre? The derivation is very straightforward. – MathematicsStudent1122 Feb 07 '16 at 01:06
  • Usually derivatives do not exist when there are "gaps" in the function. In this case for the absolute value of $x^x$ has a derivative can exist in clustered points but only if it coincides with one function. I found for $(-2)^{x}$ since it coincides with two functions its derivative does not exist which can be shown in http://math.stackexchange.com/questions/1505310/we-can-define-the-derivative-of-a-function-whose-domain-is-a-subset-of-rational – Arbuja Feb 07 '16 at 01:17
  • 1
    I was referring to the positive domain. I wasn't aware of its behaviour for $x<0$. Interesting, thanks for the link! – MathematicsStudent1122 Feb 07 '16 at 02:03
  • 1
    Have you seen this answer on concerning the $x^x$ spindle? If you choose the principal branch, then you can use ordinary integration to get an average value of $0.00704628 - 0.0121195i$. – Mark McClure Feb 07 '16 at 02:42
  • I know I can convert this into complex integral form but in my original question I mentioned about finding the average of defined points of $x^x$ when $x<0$. This is in the set $x=\frac{m}{2k+1}$ where $m$ and $k$ are negative integers. – Arbuja Feb 07 '16 at 02:50
  • @MarkMcClure I am talking about real numbers. – Arbuja Feb 07 '16 at 03:05
  • my hint: function $x^x$ has a minimum value at $x=\frac 1e$ – Bhaskara-III Feb 07 '16 at 23:07
  • @Bhaskara-III I am looking at the negative domain not the positive. Read my post below my question and if does not make sense let me know. – Arbuja Feb 07 '16 at 23:10
  • I think I know what I am doing but I need someone to check my work. – Arbuja Feb 09 '16 at 20:41
  • 1
    As far as I know, a Riemannian integral of $f$ requires $f$ to be defined on an interval. Not just on a countably infinite set of points within the interval, but on all of the interval. Your real-valued function $x^x$ is not defined on any interval within $[-4,-2]$, so there's not even a question about which intervals to choose: there are none to choose from. I don't see how to do this with integration at all. – David K Feb 14 '16 at 19:06
  • 1
    If you constrain the input to a finite set, e.g. $x \in [-4,-2]$ such that $x =\frac m{2k+1}$, for $m$, $k$ integers with $k < N$, you can take an average, and maybe it's interesting to see what happens if you let $N$ tend to infinity. I'm not sure you can call this "the" average of $f$ on $[-4,-2]$, but at least it might be something you can actually define. – David K Feb 14 '16 at 19:09
  • @DavidK Could there be some technique other than integration? – Arbuja Feb 14 '16 at 22:56
  • @DavidK I already took the average of this at $x=\frac{m}{2k+1}$ as shown in my post. In the summation form of the integral (for the constraint) ($\frac{1}{2}\lim_{n\to\infty}\sum_{i=1}^{n}{f\left(-4+\left(\frac{2}{n}\right)i\right)}\left(\frac{2}{n}\right)$), if I choose 2*(An odd integer) $n\to\infty$ I get zero. If I choose an odd integer $n$ I get $.062152$. Since there are undefined gaps I think that interval should be as dense as possible leading me to believe the answer is zero. – Arbuja Feb 16 '16 at 20:12

1 Answers1

2

There is at least one sense in which you might say the "average" value of $f(x) = x^x$ on $[-4,-2]$ is zero.

Before getting to that, however, I'd like to document some of the ways not to approach this problem. For various reasons, I think trying to apply integration is a non-starter. The function as we understand it here is defined only on a countable number of points (rational exponents) and undefined on an uncountable number of points in the interval. This makes it impossible to apply the definition of Riemannian integration (in any form I've ever seen it), and the Lebesgue integral of any function over a countable set of points is zero by definition regardless of the function values, so it tells us nothing about the average value of the function.

In addition to not having a good domain for integration, within that domain you have function values that are greater than $1/256$ interleaved densely with function values that are less than $-1/256$. So any time you use one of those values as the height of a rectangle "approximating" the function, it's completely unrepresentative of infinitely many function values in the interval. If you attempt to define an integral using only Riemann sums of $n$ rectangles with uniform widths (which is not a legitimate use of Riemann sums in the first place), as $n$ goes to infinity, each time $n$ increases you'll have some number of intervals on which that interval's part of its "Riemann" rectangle "flips" across the axis. This process never converges. Worse still, whenever $n$ is divisible by $4$, about half the values of $x^x$ for $x = -4 + j\frac2n$ are not even defined, because after reducing the fraction to lowest terms, the denominator is even.

So basically I would immediately discard any ideas that would define the average in terms of an integral, or anything purporting to be some kind of Riemann sum.

The next best thing would be if we could somehow average the function values as an infinite series. There are only a countable number of $x$ values in the domain of this function, so they can be organized in a series. Unfortunately for this approach, a series using these function values cannot be absolutely convergent. Even after we divide each partial sum by the number of terms, it's still possible to "converge" to any sum you want by choosing the order in which the terms are added.

One thing we can do is (somewhat arbitrarily) decide that we're interested in some particularly "regular" subsequences of the function values, for example, for some integer $k$, take $x^x$ where $x$ ranges over all multiples of the fraction $\frac{1}{2k+1}$ within the interval $[-4,-2]$. Take the average of these values, and then see what happens to that average as $k$ goes to infinity; that is, for $f(x) = x^x$, define $$ \mu = \lim_{k\to\infty} \frac{1}{4k+3} \sum_{j=0}^{4k+2} f \left(-4 + \frac{j}{2k+1} \right) $$ and call $\mu$ defined in this way a kind of "average".

(Note: Despite its resemblance to Riemann sums that are sometimes used when computing Riemann integrals, this is not the derivation of an integral. It is simply an arithmetic mean of a set of values of $f(x)$ for $x \in [-4,-2]$. To actually have a Riemann integral, you have to consider all partitions of the interval into "almost disjoint" subintervals, including many for which it is not possible to construct a Riemann sum over regularly spaced values of $x$ like these. And let's not forget that rigorous definitions of Riemann integration typically assume a function that is defined on a compact interval, not a function that is undefined on a dense subset of points in that interval.)

The value of this "average" is zero. Here's how we can know that:

For any integer $k$, we can separate the aforementioned sum into four parts: the first term, the last term, the "odd" terms in between, and the "even" terms in between: \begin{split} \sum_{j=0}^{4k+2} f \left(-4 + \frac{j}{2k+1} \right) = f(-4) &+ \sum_{j=0}^{2k} f \left(-4 + \frac{2j+1}{2k+1} \right) \\ &+ \sum_{j=1}^{2k} f \left(-4 + \frac{2j}{2k+1} \right) + f(-2). \end{split}

Define a function $g$ by $g(x) = x^{-x}$. If $x = -\frac pq$, then for $p$ even and $q$ odd, $$f(x) = \left(-\frac pq\right)^{-p/q} = \left(\frac pq\right)^{-p/q} = g(-x),$$ but for $p$ and $q$ both odd, $$f(x) = \left(-\frac pq\right)^{-p/q} = -\left(\frac pq\right)^{-p/q} = -g(-x).$$ So we can rewrite the sum as \begin{split} \sum_{j=0}^{4k+2} f \left(-4 + \frac{j}{2k+1} \right) = g(4) &- \sum_{j=0}^{2k} g \left(4 - \frac{2j+1}{2k+1} \right) \\ &+ \sum_{j=1}^{2k} g \left(4 - \frac{2j}{2k+1} \right) + g(2) \\ = g(2) &- \sum_{j=0}^{2k} g \left(2 + \frac{2j+1}{2k+1} \right) \\ &+ \sum_{j=1}^{2k} g \left(2 + \frac{2j}{2k+1} \right) + g(4). \end{split}

Since $g$ is a decreasing function over $[2,4]$, $$ g \left(2 + \frac{2j-1}{2k+1} \right) > g \left(2 + \frac{2j}{2k+1} \right) > g \left(2 + \frac{2j+1}{2k+1} \right), $$ so we add the term $g(2)$ to the sum of the "even" terms and compare them to the sum of the "odd" terms, we find that $$ g(2) + \sum_{j=1}^{2k} g \left(2 + \frac{2j}{2k+1} \right) > \sum_{j=0}^{2k} g \left(2 + \frac{2j+1}{2k+1} \right), $$ but if we instead add the sum of the "even" terms to $g(4)$, we find that $$ \sum_{j=0}^{2k} g \left(2 + \frac{2j+1}{2k+1} \right) > \sum_{j=1}^{2k} g \left(2 + \frac{2j}{2k+1} \right) + g(4). $$ We can conclude that $$ -g(2) < \sum_{j=1}^{2k} g \left(2 + \frac{2j}{2k+1} \right) - \sum_{j=0}^{2k} g \left(2 + \frac{2j+1}{2k+1} \right) < -g(4). $$ and therefore $$ \frac{1}{256} = (g(2) + g(4)) - g(2) < \sum_{j=0}^{4k+2} f \left(-4 + \frac{j}{2k+1} \right) < (g(2) + g(4)) - g(4) = \frac14. $$ Since the sum is bounded for any $k$, $$ \mu = \lim_{k\to\infty} \frac{1}{4k+3} \sum_{j=0}^{4k+2} f \left(-4 + \frac{j}{2k+1} \right) = 0 $$

Now, to make this just a little more interesting, instead of taking just the multiples of $\frac{1}{2k+1}$, suppose we consider all multiples of any fraction $\frac{1}{2t+1}$ for any integer $t \leq k$, but count each such rational number only once. That is, instead of throwing away all the multiples of $\frac13$ and replacing them with multiples of $\frac15$, then throwing them away and replacing them with multiples of $\frac17$, we let the smaller fractions "fill in" the gaps between the larger ones without throwing anything away. So at $k=3$ we have terms in the sum for all multiples of $\frac13$, $\frac15$, and $\frac17$ in the interval $[-4,-2]$, but the values $f(-4)$, $f(-3)$, and $f(-2)$ are all counted only once in the sum. For $k=4$ we add terms for all the multiples of $\frac19$ except the multiples of $\frac13$, since we already have those; and so forth.

This would be a relatively satisfying limit to evaluate. I suspect that the limit of the sum as $k$ goes to infinity is still zero, although I do not have a proof at this time. (Each new batch of terms resembles one of the $\frac{1}{2k+1}$ sums from the proof above, but it has "holes" that are not accounted for in that proof; otherwise this new limit would be practically a corollary of the previous one.)

David K
  • 98,388
  • Are the any textbook which state that riemman integral cannot exist if undefined on an uncountable number of points? – Arbuja Sep 18 '16 at 13:59
  • If we instead took $|x^x|$ which is defined on $x=\left{\frac{m}{2n+1}|m,n \in \mathbb{Z}\right}$ but coincides with only $(-x)^x$ then could a riemman integral exist? – Arbuja Oct 09 '16 at 16:33
  • Considering the density of points with one function if we could approximate arbirarily close to any $(-{t_i})^{t_i}$ in $\sum_{i=1}^{n}f(t_i)(x_{i+1}-x_i)$ we would have a definite integral on all real numbers. – Arbuja Oct 09 '16 at 16:39
  • The point of Riemann integration is that you don't consider just some particularly "nice" partitions of the interval of integration that give you the terms you want to see in your Riemann sum. You also have to get the same result from all kinds of partitions of the interval. And of course a rigorous definition of a Riemann integral will usually start by assuming the function is defined on the interval, which immediately disqualifies $x^x$ over $[-4,-2]$. – David K Oct 09 '16 at 17:23
  • For example, at the bottom of page 2 of https://www.math.ucdavis.edu/~hunter/m125b/ch1.pdf : "Suppose that $f:[a,b]\to\mathbb R$ is a bounded function on the compact interval $I=[a,b]$ ...". The expression $x^x$ is not a function on $[-4,-2]$, because to be a function on an interval you have to assign a function value to every point in the interval. – David K Oct 09 '16 at 17:31
  • A particular problem with this function is that you have positive values (not near zero) and negative values (also not near zero) both dense within every subinterval of the desired interval of integration, so your lower Riemann integral will never equal your upper Riemann integral no matter how you "fill the gaps" in the function. Really, I would not worry too much about whether the limit of the sums is an integral. It is possible to have such a limit that is not (strictly speaking) an integral but is still interesting and "like an integral." – David K Oct 09 '16 at 19:27
  • But if we took the absolute value of $x^x$ then there is only positive values that are defined on $x=\left{\frac{m}{2n+1}|m,n \in \mathbb{Z}\right}$. So then could a reimmen integral exist? – Arbuja Oct 09 '16 at 19:50
  • My hypothesis is that the definite integral of $|x^x|$ would be the same as the definite integral of $(-x)^{x}$ – Arbuja Oct 09 '16 at 19:56
  • Right?${}{}{}{}$ – Arbuja Oct 14 '16 at 18:25
  • 1
    I think it's true that $(-x)^x = \lvert x^x\rvert$ whenever $x<0$ and $\lvert x^x\rvert$ is real. Also, $(-x)^x$ is continuous for $x<0$, its integral is defined, and $\int_{-4}^{-2} (-x)^x,dx = \int_2^4 x^{-x},dx \approx 0.130243$. I don't think that's really an integral of $\lvert x^x\rvert$, but it is an integral of a continuous extension of $\lvert x^x\rvert$. I still think the sums in your original question are more interesting. – David K Oct 14 '16 at 20:32