## triples (x,y,z) with particular properties

For the discussion of math. Duh.

Moderators: gmalivuk, Moderators General, Prelates

alessandro95
Posts: 109
Joined: Wed Apr 24, 2013 1:33 am UTC

### triples (x,y,z) with particular properties

Hello everyone, this problem is taken from the entrance test of an Italian university famous for being extremely hard, this is the first problem of the 13 pages long entrance test and I already have no idea on what to do with it (no, I'm not planning to even try this test!):

the original text for Italian speakers is:
"Si trovino le terne di numeri reali (x,y,z) con la proprieta' che la quarta potenza di ciascuno di essi e' uguale alla somma degli altri due."

That translates as:
"Find the triples of real numers (x,y,z) with the property that the fourth power of each of them is equal to the sum of the other two"

which is equivalent to the system of 3 equations:
x^4=y+z
y^4=x+z
z^4=y+x

but I cannot do anything better than pointing out the trivial solution for (0,0,0), any hint on how to solve this problem?
Thank in advance for any help!
The primary reason Bourbaki stopped writing books was the realization that Lang was one single person.

vbkid
Posts: 41
Joined: Wed Oct 31, 2012 5:53 pm UTC

### Re: triples (x,y,z) with particular properties

alessandro95 wrote:Hello everyone, this problem is taken from the entrance test of an Italian university famous for being extremely hard, this is the first problem of the 13 pages long entrance test and I already have no idea on what to do with it (no, I'm not planning to even try this test!):

the original text for Italian speakers is:
"Si trovino le terne di numeri reali (x,y,z) con la proprieta' che la quarta potenza di ciascuno di essi e' uguale alla somma degli altri due."

That translates as:
"Find the triples of real numers (x,y,z) with the property that the fourth power of each of them is equal to the sum of the other two"

which is equivalent to the system of 3 equations:
x^4=y+z
y^4=x+z
z^4=y+x

but I cannot do anything better than pointing out the trivial solution for (0,0,0), any hint on how to solve this problem?
Thank in advance for any help!

I just observed that x=y=z So:
x^4=2x
x^3=2
x=y=z=2^(1/3)

alessandro95
Posts: 109
Joined: Wed Apr 24, 2013 1:33 am UTC

### Re: triples (x,y,z) with particular properties

yes, I also found 2^(1/3) as solution just after posting this thread, how do you know that x,y and z must all be equal to each other?
The primary reason Bourbaki stopped writing books was the realization that Lang was one single person.

faldion
Posts: 8
Joined: Sun Mar 01, 2009 12:21 pm UTC

### Re: triples (x,y,z) with particular properties

If x ≠ y, we can do the following:

x^4 = y + z
y^4 = x + z
Subtract the lower equation from the upper one to get x^4 - y^4 = y - x.

x^4 - y^4 = y - x
(x^2 + y^2)(x^2 - y^2) = y - x
(x^2 + y^2)(x - y)(x + y) = -(x - y), divide by (x - y) to get
(x^2 + y^2)(x + y) = -1

(x^2 + y^2) is positive for real x and y, so (x + y) is necessarily negative because their product is negative. But (x + y) = z^4, which is positive for real z, and we have a contradiction.

Therefore x = y. A similar argument works for y = z, and we're done.

alessandro95
Posts: 109
Joined: Wed Apr 24, 2013 1:33 am UTC

### Re: triples (x,y,z) with particular properties

you guys are awesome, thank a lot!
It looked harder to me, well it still is the first of a lot of exercise after all...!

I may want to post some other exercise from the same test, after thinking a bit longer, should I open a new thread or post here?
The primary reason Bourbaki stopped writing books was the realization that Lang was one single person.

vbkid
Posts: 41
Joined: Wed Oct 31, 2012 5:53 pm UTC

### Re: triples (x,y,z) with particular properties

Mine was intuition (whether it was correct or not is another question!) but the algebra above brought me to the same conclusion.
Thanks for this post, it was a good, quick stretch of the brain!

vbkid
Posts: 41
Joined: Wed Oct 31, 2012 5:53 pm UTC

### Re: triples (x,y,z) with particular properties

I'd say just post here!

jestingrabbit
Factoids are just Datas that haven't grown up yet
Posts: 5967
Joined: Tue Nov 28, 2006 9:50 pm UTC
Location: Sydney

### Re: triples (x,y,z) with particular properties

vbkid wrote:I just observed that x=y=z So:
x^4=2x
x^3=2
x=y=z=2^(1/3)

You're forgetting the rather tedious x = y = z = 0 solution.
ameretrifle wrote:Magic space feudalism is therefore a viable idea.

vbkid
Posts: 41
Joined: Wed Oct 31, 2012 5:53 pm UTC

### Re: triples (x,y,z) with particular properties

jestingrabbit wrote:
vbkid wrote:I just observed that x=y=z So:
x^4=2x
x^3=2
x=y=z=2^(1/3)

You're forgetting the rather tedious x = y = z = 0 solution.

Thw OP actually covered it already.

alessandro95
Posts: 109
Joined: Wed Apr 24, 2013 1:33 am UTC

### Re: triples (x,y,z) with particular properties

vbkid wrote:Mine was intuition (whether it was correct or not is another question!) but the algebra above brought me to the same conclusion.
Thanks for this post, it was a good, quick stretch of the brain!

The problem seems to ask only for the values, not for a demonstration that these are the only solutions, so your answer would have probably been enough for the test ok, next problem, I already did some work on it (spoilered):
"Consider an integer n and the number Sn of increasing sequences of integer starting from 0 and ending at n, alternatively even and odd, for example S3=2 since the only 2 such sequences are 0,1,2,3 and 0,3.'
Consider the Fibonacci sequence:
f1=f2=1
fn=fn-1+fn-2, n>2
Prove that Sn=fn

Spoiler:
I started writing by hand the first values of Sn
f1=1 (0,1)
f2=1 (0,1,2)
f3=2 (0,1,2,3)(0,3)
f4=3 (0,1,2,3,4)(0,1,4)(0,3,4)
f5=5 (0,1,2,3,4,5)(0,1,4,5)(0,3,4,5)(0,1,2,5)(0,5)
f6=8 (0,1,2,3,4,5,6)(0,1,4,5,6)(0,3,4,5,6)(0,1,2,5,6)(0,5,6)(0,1,6)(0,1,2,3,6)(0,3,6)

To make the sequences for an even n we only need to append n to all the sequences with an odd index smaller than n, for example to make S4 we append 4 to the sequences for S1 and S3.
In a similar way to make the sequences for an odd Sn we append n to the sequences with an even smaller index and then we add the sequence (0,n) that cannot be formed in this way, for example to make S5 we append 5 to the sequences for S2 and S4 and then we also add the sequence (0,5).

Which is equivalent to say that: and which, I guess, is a well known property of the Fibonacci sequence, tough I have no idea on how to demonstrate it, any hints?

Please use spoilers if you post a whole solution!

p.s.
how do I use the [imath] tag? I had to use and external LaTeX implementation
Last edited by alessandro95 on Thu May 16, 2013 4:06 am UTC, edited 1 time in total.
The primary reason Bourbaki stopped writing books was the realization that Lang was one single person.

DeGuerre
Posts: 51
Joined: Mon Feb 04, 2008 6:41 am UTC

### Re: triples (x,y,z) with particular properties

A comment on your spoiler, which I am posting here not in a spoiler, because it doesn't give anything away:

alessandro95 wrote:which, I guess, is a well known property of the Fibonacci sequence, tough I have no idea on how to demonstrate it, any hints?

You're going to be embarrassed by how obvious this hint is.

f(2n) = f(2n-1) + f(2n-2) = f(2n-1) + f(2(n-1))

jedelmania
Posts: 75
Joined: Tue Jan 15, 2013 5:48 pm UTC

### Re: triples (x,y,z) with particular properties

alessandro95 wrote:To make the sequences for an even n we only need to append n to all the sequences with an odd index smaller than n, for example to make S4 we append 4 to the sequences for S1 and S3.

This is correct, and you get there from the definition of the Fibonacci sequence

Hint
Spoiler:
write out the definition of the series f(k) = f(k-1) + f(k-2) + use induction

There is a quicker way to the answer
Spoiler:
To make the sequences for n we only need to append n to all the sequences from Sn-1 and replace n-2 with n from all the sequences of Sn-2. Proof follows immediately from this observation.