My colleague Jason Ermer at Collaborative Mathematics is back from summer hiatus and has published Challenge 16 on his website: http://www.collaborativemathematics.org/challenge16.html
My colleague Jason Ermer at Collaborative Mathematics is back from summer hiatus and has published Challenge 16 on his website: http://www.collaborativemathematics.org/challenge16.html
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
As we’ve seen in this series, the answer is
Also, as long as and
, there is an alternative answer:
.
In this concluding post of this series, I’d like to talk about the practical implications of the assumptions that and
.
For the sake of simplicity for the rest of this post, let
and
.
If I evaluate a definite integral of over an interval that contains neither
or
, then either
or
can be used. Courtesy of Mathematica:
However, if the region of integration contains either or
(or both), then only using
returns the correct answer.
So this should be a cautionary tale about solving for angles, as the innocent-looking that appeared several posts ago ultimately makes a big difference in the final answers that are obtained.
In the course of evaluating the antiderivative
,
I have stumbled across a very curious trigonometric identity:
if
,
if
,
if
,
where and
are the unique values so that
,
.
I will now show that and
. Indeed, it’s apparent that these have to be the two transition points because these are the points where
is undefined. However, it would be more convincing to show this directly.
To show that , I need to show that
.
I could do this with a calculator…
Instead, let and
, so that
,
.
Indeed, by SOHCAHTOA, the angles and
can be represented in the figure below:
The two small right triangles make one large triangle, and I will show that the large triangle is also a right triangle. To do this, let’s find the lengths of the three sides of the large triangle. The length of the longest side is clearly . I will use the Pythagorean theorem to find the lengths of the other two sides. For the small right triangle containing
, the missing side is
Next, for the small right triangle containing , the missing side is
So let me redraw the figure, eliminating the altitude from the previous figure:
Notice that the condition of the Pythagorean theorem is satisfied, since
,
or
.
Therefore, by the converse of the Pythagorean theorem, the above figure must be a right triangle (albeit a right triangle with sides of unusual length), and so . In other words,
, as required.
To show that , I will show that the function
is an odd function using the fact that
is also an odd function:
.
Therefore, , and so
.
In the course of evaluating the antiderivative
,
I’ve accidentally stumbled on a very curious looking trigonometric identity:
if
,
if
,
if
.
The extra and
are important. Without them, the graphs of the left-hand side and right-hand sides are clearly different if
or
:
However, they match when those constants are included:
Let’s see if I can explain why this trigonometric identity occurs without resorting to the graphs.
Since assumes values between
and
, I know that
,
,
and so
.
However,
,
and so and
must differ if
is in the interval
or in the interval
.
I also notice that
,
,
and so
.
However, this difference can only be equal to a multiple of , and there are only three multiples of
in the interval
, namely
,
, and
.
To determine the values of where this happens, I also note that
,
, and
are increasing functions, and so
must also be an increasing function. Therefore, to determine where
lies in the interval
,it suffices to determine the unique value
so that
. Likewise, to determine where
lies in the interval
,it suffices to determine the unique value
so that
.
In summary, I have shown so far that
if
,
if
,
if
,
where and
are the unique values so that
,
.
So, to complete the proof of the trigonometric identity, I need to show that and
. I will do this in tomorrow’s post.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
As we’ve seen in this series, the answer is
It turns out that this can be simplified somewhat as long as and
. I’ll use the trig identity
When I apply this trig identity for and
, I obtain
.
So we can conclude that
for some integer that depends on
. The
is important, as a cursory look reveals that
and
have different graphs. (The vertical lines in the orange graph indicate where the right-hand side is undefined when
or
.
The two graphs coincide when but differ otherwise. However, it appears that the two graphs differ by a constant. Indeed, if I subtract
from the orange graph if
and add
to the orange graph if
, then they match:
So, evidently
if
,
if
,
if
.
So as long as and
, this constant
,
, or
can be absorbed into the constant
:
.
However, a picture may be persuasive but is not a proof, and there are some subtle issues with this simplification. I’ll discuss these further details in tomorrow’s post.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
So far, I’ve shown that the denominator can be factored over the real numbers:
To evaluate the remaining two integrals, I’ll use the antiderivative
.
To begin, I’ll complete the squares:
Applying the substitutions and
, I can continue:
Combining, I finally arrive at the answer for :
Naturally, this can be checked by differentiation, but I’m not going type that out.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
So far, I’ve shown that the denominator can be factored over the real numbers:
after finding the partial fractions decomposition.
Let me start with the first of the two integrals. It’d be nice to use the substitution . However,
, and so this substitution can’t be used cleanly. So, let me force the numerator to have this form, at least in part:
The substitution can now be applied to the first integral:
.
On the last line, I was able to remove the absolute value signs because is an irreducible quadratic and hence is never equal to zero for any real number
.
Similarly, I’ll try to apply the substitution to the second integral:
The substitution can now be applied to the first integral:
.
So, thus far, I have shown that
I’ll consider the evaluation of the remaining two integrals in tomorrow’s post.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
So far, I’ve shown that the denominator can be factored over the real numbers:
,
so that the technique of partial fractions can be applied. Since both quadratics in the denominator are irreducible (and the degree of the numerator is less than the degree of the denominator), the partial fractions decomposition has the form
Clearing out the denominators, I get
or
or
Matching coefficients yields the following system of four equations in four unknowns:
Ordinarily, four-by-four systems of linear equations are somewhat painful to solve, but this system isn’t too bad. Since from the first equation, the third equation becomes
, or
.
From the fourth equation, I can conclude that and
. The second and third equations then become
,
or
,
.
Adding the two equations yields , so that
and
.
Therefore, the integral can be rewritten as
I’ll start evaluating this integral in tomorrow’s post.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
To compute this integral, I will use the technique of partial fractions. In yesterday’s post, I used De Moivre’s Theorem to factor the denominator over the complex plane, which then led to the factorization of the denominator over the real numbers.
In today’s post, I present an alternative way of factoring the denominator by completing the square. However, unlike the ordinary method of completing the square, I’ll do this by adding and subtracting the middle term and not the final term:
.
The quadratic formula can then be used to confirm that both of these quadratics have complex roots and hence are irreducible over the real numbers, and so I have thus factored the denominator over the real numbers:
.
and the technique of partial fractions can be applied.
There’s a theorem that says that any polynomial over the real numbers can be factored over the real numbers using linear terms and irreducible quadratic terms. However, as seen in this example, there’s no promise that the terms will have rational coefficients.
I’ll continue the calculation of this integral with tomorrow’s post.
This antiderivative has arguable the highest ratio of “really hard to compute” to “really easy to write”:
To compute this integral, I will use the technique of partial fractions. This requires factoring the denominator over the real numbers, which can be accomplished by finding the roots of the denominator. In other words, I need to solve
,
or
.
I switched to the letter since the roots will be complex. The four roots of this quartic equation can be found with De Moivre’s Theorem by writing
,
where is a real number, and
By De Moivre’s Theorem, I obtain
.
Matching terms, I obtain the two equations
and
or
and
or
and
.
This yields the four solutions
Therefore, the denominator can be written as the following product of linear factors over the complex plane:
or
or
or
or
.
We have thus factored the denominator over the real numbers:
,
and the technique of partial fractions can be applied.
I’ll continue the calculation of this integral with tomorrow’s post.