Functional Equation Fun

Algebra Level 5

f ( x ) + f ( 6 x 5 4 x 2 ) = x f(x)+f\left(\frac{6x-5}{4x-2}\right)=x

Find the number of real -valued functions f f that are solutions of the above functional equation and whose domain is the set of all real numbers different from 1 2 \frac{1}{2} , 1 1 , and 3 2 \frac{3}{2} .


The answer is 0.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

4 solutions

Arturo Presa
May 30, 2016

Relevant wiki: Functional Equations - Problem Solving

We are going to use the method of contradiction to prove that the given equation does not have solutions on the domain described in the question. Let us assume that such a function f f exists and ϕ ( x ) = 6 x 5 4 x 2 . \phi (x)= \frac{6x-5}{4x-2}. By using the notation ϕ n \phi^n for the composition of ϕ \phi with itself n n times, we obtain that ϕ 2 ( x ) = 4 x 5 4 x 4 , \phi ^2(x)=\frac{4x-5}{4x-4}, ϕ 3 ( x ) = 2 x 5 4 x 6 , \phi ^3(x)=\frac{2x-5}{4x-6}, and ϕ 4 ( x ) = x , \phi ^4(x)=x, for all values of x x in the domain of f f . We can also see that all the values of the functions ϕ ( x ) , \phi (x), ϕ 2 ( x ) , \phi^2 (x), ϕ 3 ( x ) , \phi^3 (x), and ϕ 4 ( x ) = x \phi^4(x)=x are in the domain of f . f. Substituting into the functional equation x x by ϕ ( x ) , \phi (x), ϕ 2 ( x ) , \phi^2 (x), and ϕ 3 ( x ) , \phi^3 (x), respectively, we get three new equations that together with the original one form the following system: f ( x ) + f ( ϕ ( x ) ) = x f ( ϕ ( x ) ) + f ( ϕ 2 ( x ) ) = ϕ ( x ) f ( ϕ 2 ( x ) ) + f ( ϕ 3 ( x ) ) = ϕ 2 ( x ) f ( ϕ 3 ( x ) ) + f ( x ) = ϕ 3 ( x ) \begin{aligned} f(x)+f(\phi(x)) & =x \\ f(\phi(x))+f(\phi^2(x))& =\phi(x)\\ f(\phi^2(x))+f(\phi^3(x))& =\phi^2(x)\\ f(\phi^3(x))+f(x)& =\phi^3(x) \end{aligned} Now,you can notice that for any x x in the domain of f f the following is true: f ( ϕ 3 ( x ) ) + f ( x ) = f ( x ) + f ( ϕ ( x ) ) ( f ( ϕ ( x ) ) + f ( ϕ 2 ( x ) ) ) + f ( ϕ 2 ( x ) ) + f ( ϕ 3 ( x ) ) = x ϕ ( x ) + ϕ 2 ( x ) ϕ 3 ( x ) . f(\phi^3(x))+f(x)=f(x)+f(\phi(x)) -(f(\phi(x))+f(\phi^2(x)))+f(\phi^2(x))+f(\phi^3(x))=x-\phi (x)+\phi^2 (x)\neq\phi^3 (x). This is the contradiction, so our proof is complete.

Did the same. A very nice question and solution indeed

Aditya Kumar - 5 years ago

Log in to reply

Thank you, Aditya!

Arturo Presa - 5 years ago

Very nice sol.n sir +1!!. My approach was a bit logical & in a similar method.

Rishabh Tiwari - 5 years ago

Log in to reply

Thank you, Rishabh!

Arturo Presa - 5 years ago

It is just that it is so hard to find that this is cyclic.

Ruilin Wang - 1 year, 10 months ago

I did the same untill the system of 4 equations. Then I added up the first plus third equation and second plus fourth equation. Then I noticed that RHSs of the two new equations are not the same so contradiction. And so I was wondering this functional equation must have 0 or ∞ solutions of the cycle is 4 because either the two new RHSs are equal for all x, or not. Is that true?

Veselin Dimov - 6 months, 2 weeks ago

Please try my problem sir; Hope u like it:)

Rishabh Tiwari - 5 years ago
Otto Bretscher
May 30, 2016

Compañero Presa has written a careful systematic solution (+1), as usual. Being in a hurry, I did the problem by considering an example:

f ( 2 ) + f ( 7 / 6 ) = 2 , f(2)+f(7/6)=2, f ( 7 / 6 ) + f ( 3 / 4 ) = 7 / 6 , f ( 3 / 4 ) + f ( 1 / 2 ) = 3 / 4 , f ( 1 / 2 ) + f ( 2 ) = 1 / 2 f(7/6)+f(3/4)=7/6,f(3/4)+f(-1/2)=3/4, f(-1/2)+f(2)=-1/2 . If we add the first and third equation and subtract the second and fourth, we find 0 = 25 12 0=\frac{25}{12} , a contradiction. Thus no such function exists.

How did you know you should substitute x = 2 x = 2 at the very beginning?

Pi Han Goh - 5 years ago

Log in to reply

I picked "the simplest number in the domain" (1 is excluded). I suspected that the thing would have to be periodic... otherwise the problem would have infinitely many solutions, an option that is not offered.

Otto Bretscher - 5 years ago

Log in to reply

Your proof is similar as mine. I searched f ( 0 ) f(0) and I got a contradiction. My general method for these kind of problems,when there aren't extra conditions (continuous, differentiable..)is firstly to see what happens with natural numbers,starting generally with 0 and 1( in this case 1 is not an option), later to see what happens with integers and rational numbers, and finally if I haven't still found the solution to see what happens with real numbers... ( I generally use also this method when the only condition is continuity...) For example, I did this(to look for f ( 0 ) f(0) ) here and here . Of course, I consider Arturo's solution more elegant than mine, although we eventually got the same conclusion. Once, a clever buddy told me when you try to prove something and you can't get it directly, suppose otherwise that it is not true, and try to see what happens... in other words, use reductio ad absurdum... In this case, it hasn't been necessary for me, but other times I use it...

Guillermo Templado - 5 years ago

Log in to reply

@Guillermo Templado As Compañero Presa mentions, there is something to be said for the "economy of means." Why do the problem in general when considering a simple case (like 0 or 2) suffices?

Otto Bretscher - 5 years ago

Log in to reply

@Otto Bretscher In this case( 0 or 2) ,it has given us the solution. In other cases, it can reveal many things about the function if we know its behavior with natural numbers... I think

Guillermo Templado - 5 years ago

Log in to reply

@Guillermo Templado Exactly, Compañero!

Otto Bretscher - 5 years ago

@Otto Bretscher I agree with both of you, Compañero Otto, and Guillermo. I think the situation would have been a little different if the question were to prove that there was no such a solution on any possible domain. Then in this case to find a contradiction just using one number would not be enough. What do you think?

Arturo Presa - 5 years ago

Log in to reply

@Arturo Presa Let's suppose that a teacher(or a person) ask us for proving a proposition which it is true. We have 3 options:

1.- We can try to prove the proposition directly using prior knowledge.

2.- If this doesn't work, we have two options more, which my friend meant:

a) Use reductio ad absurdum, I mean suppose that the proposition is false and try to find a contradiction, and the last one and more intereresting, which it has been used for me and Otto in this problem. I recognize i haven't notice until the end...

b) Suposse that the proposition is true and it's, in fact, true. What can we infer then?...It can give us the path(way) to the proof from back to the front. And this has been the case here.. Otto and I have supposed than at least there is one function fullfiling the requesites and it has given us the answer.

Guillermo Templado - 5 years ago

Log in to reply

@Guillermo Templado I did not even think of it that deeply. My attitude was this: Let's see what's going on here, let's play around with some numbers. I was pleasantly surprised when I noticed that I was done after just a few minutes of fun.

PS: I want to let you Compañeros know that I'm flying to Europe today, and I will be traveling all over the place during the next two weeks. I don't expect to be on Brilliant much during this time, but I'll be back.

Otto Bretscher - 5 years ago

Log in to reply

@Otto Bretscher Yes, it has been like this... Anyway, Arturo, very good problem, and as usual interesting \uparrow

P.S.- Otto, come back soon to Brilliant, please...I'll miss you

Guillermo Templado - 5 years ago

Log in to reply

@Guillermo Templado Thank you, Guillermo, and compañero Otto, enjoy your travel!

Arturo Presa - 5 years ago

Compañero Bretscher, you always come out with the simplest, most elegant solution!

Arturo Presa - 5 years ago

Log in to reply

I'm not sure my solution is elegant. Funny thing happened: I started out by doing some numerical experimentation, to get a "feel" for the problem. After the first trial I realized that I was done!

Otto Bretscher - 5 years ago

Log in to reply

I consider your proof elegant, because of the economy of means. But it is amazing how fast you picked the essence of the question.

Arturo Presa - 5 years ago

I used this type of reasoning but with different values of x...but really awesome thinking sir...

Istiak Reza - 4 years, 12 months ago
Ángela Flores
Jul 17, 2016

We can suppose that a function like that exists. In that case, consider x = 0 , 5 2 , 5 4 , 5 6 x=0, \frac{5}{2}, \frac{5}{4},\frac{5}{6} in the domain of f. We obtain the following system: f ( 0 ) + f ( 5 2 ) = 0 f(0)+f(\frac{5}{2})=0 ; f ( 5 2 ) + f ( 5 4 ) = 5 2 f(\frac{5}{2})+f(\frac{5}{4})=\frac{5}{2} ; f ( 5 4 ) + f ( 5 6 ) = 5 4 f(\frac{5}{4})+f(\frac{5}{6})=\frac{5}{4} ; f ( 5 6 ) + f ( 0 ) = 5 6 f(\frac{5}{6})+f(0)=\frac{5}{6} . The system has no solution, therefore, the function is not well defined for 4 values in the domain of f, a contradiction.

Arousse Fares
Sep 13, 2018

I found the same contradiction as Angela's ( the way I stumbled on it is prettier however !!).

f ( 0 ) = f ( 5 2 ) f(0)=-f(\frac{5}{2})

= 5 2 + f ( 5 4 ) =-\frac{5}{2} + f(\frac{5}{4})

= 5 2 + 5 4 f ( 5 6 ) =-\frac{5}{2} + \frac{5}{4} - f(\frac{5}{6})

= 5 2 + 5 4 5 6 + f ( 0 ) =-\frac{5}{2} + \frac{5}{4} - \frac{5}{6} + f(0)

Morality : Recycling is not a waste of time.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...