Random Functional Equations

by utkarshgupta, Mar 4, 2016, 12:34 PM

Problem : (here)
Find all functions$ f\colon \mathbb{R} \to \mathbb{R} $
$$ f(x^2 + y + f(y)) = f(x)^2 + 2y $$for all $x,y \in\mathbb{R}$

Solution :
Let $P(x,y)$ be the assertion $f(x^2+y+f(y))=f(x)^2+2y$
$f$ is obviously surjective.

Let there exist a $k \in \mathbb{R}$ such that $f(k)=0$.

$P(-k,0)$ and $P(k,0)$
$$ \implies f(-k)^2=f(k)^2$$$$\implies f(-k)=0$$
$$P(0,k) \implies f(k)=f(0)^2+k$$$$\implies k \leq 0$$
But similarly
$$P(0,-k) \implies -k \leq 0$$
Hence $k$ can only be zero.
Due to surjectivity of $f$, $\boxed{f(0)=0}$



$$P(x,0) \implies \boxed{f(x^2)=f(x)^2}$$
Thus is $t > 0$, $f(t)=f(\sqrt{t})^2 > 0$
That is $t > 0 \implies f(t) > 0$

$$P(0,y) \implies \boxed{f(y+f(y))=2y}$$
Let there exist some $b$ such that $f(b)<0$
$$P(\sqrt{-f(b)},b) \implies f(b) = f(\sqrt{-f(b)})^2+2b > 2b$$
Let $z <0$
$$f(z+f(z))=2z <0$$Using the above result,
$$2z=f(z+f(z)) > 2z+2f(z)$$

Hence we get if $z < 0 \implies f(z) <0$

From $P(x,0)-P(-x,0)$ and the result of sign (positive or negative) of $f(x)$ depending on $x$),
$$f(x)+f(-x)=0$$
Observe that
$$P(x,y) \implies f(x^2+y+f(y))=f(x^2)+2y$$$$P(x,y+f(y)) \implies f(x^2+y+f(y)+f(y+f(y))) = f(x^2)+2(y+f(y))$$$$\implies f(x^2+2y+y+f(y))=f(x^2)+2y+2f(y)$$for $x^2+2y \ge 0$
$$\implies f(x^2+2y+y+f(y))=f(x^2+2y)+2y$$
For $x^2+2y \ge 0$
Hence we get $f(x^2+2y)=f(x^2)+2f(y)$
Setting $y=-\frac{x^2}{2}$ above, we get $f(x^2)=-2f(\frac{-x^2}{2})$
This implies $f(2m)=2f(m)$

Thus $f(x^2+2y)=f(x^2)+f(2y)$

But again because of the fact $xf(x) \ge 0$ (after some case making I guess),
we get $f$ is additive that is $\boxed{f(x+y)=f(x)+f(y)}$

Also obviously by the above condition and the fact that $xf(x)>0$, the function is monotonous...

Hence by Cauchy equation, the function is linear.
Putting value $\boxed{f(x)=x}$ is the only function satisfying the condition and hence we are done.



Problem (@socrates)(here):[\b]
Determine all functions $f : \mathbb{R}^+ \to \mathbb{R}^+$ such that \[f (x + f (x + y)) = f (2x) + y,\]for all $x,y \in \mathbb{R}^+$

Solution :
Let $P(x,y)$ be the assertion $f(x+f(x+y))=f(2x)+y$

First I will show that $f$ is injective
Let $f(a)=f(b)$
We can always chose an $x \in \mathbb{R}^+$ such that $x \le a,b$
$$P(x,a-x) \implies f(x+f(a))=f(2x)+a-x$$$$P(x,b-x) \implies f(x+f(b))=f(2x)+b-x$$
That is $a=b$

Hence we have $f$ is surjective.


$$P(x,f(x+y)) \implies f(x+f(x+f(x+y)))=f(2x)+f(x+y)$$$$\implies f(x+y+f(2x))=f(2x)+f(x+y)$$Setting $2x=s, x+y=t$
$$\implies f(t+f(s))=f(s)+f(t)$$for all $t,s \in \mathbb{R}^+$ (since we can find such $x,y$
But this also implies $f(s)+f(t)=f(s+f(t))$

Thus we have
$$f(s+f(t))=f(t+f(s))$$Using injectivity and setting $s=1$
$$f(t)=t+f(1)-1$$
That is $f$ is linear.

Putting $f \equiv t+c$, we get $c=0$

Hence $\boxed{f(t)=t}$ is the only solution.
This post has been edited 2 times. Last edited by utkarshgupta, Mar 5, 2016, 1:43 AM

Comment

4 Comments

The post below has been deleted. Click to close.
This post has been deleted. Click here to see post.
Seems nice. Let me see if I can come up with a new solution. How do you do FEs so quickly? :D

by anantmudgal09, Mar 4, 2016, 1:00 PM

The post below has been deleted. Click to close.
This post has been deleted. Click here to see post.
Darn..this post reminded me that i was totally out of practice in Fes :( :P
@anant,yeah utkarsh is our indian pco :P

by kapilpavase, Mar 4, 2016, 2:51 PM

The post below has been deleted. Click to close.
This post has been deleted. Click here to see post.
Yes, maybe you should rival BJV :P

by anantmudgal09, Mar 4, 2016, 3:18 PM

The post below has been deleted. Click to close.
This post has been deleted. Click here to see post.
I don't think I do them quickly or anything.
In fact I find it rather hard to proceed on most of the ISLs...

And please pco and BJV are like ...
uk hard to describe.

by utkarshgupta, Mar 4, 2016, 4:43 PM

Stay insane,Coz it's your will, labour and pain,which takes you to the top of the mountain.

avatar

utkarshgupta
Archives
- September 2017
+ September 2016
+ July 2016
+ December 2015
+ August 2015
+ December 2014
Shouts
Submit
  • Here goes first post of 2025! Great blog.

    by math_holmes15, Jan 14, 2025, 8:53 AM

  • First post of 2024

    by Yiyj1, Feb 8, 2024, 5:40 AM

  • First post of 2023

    by HoRI_DA_GRe8, Jul 22, 2023, 7:45 AM

  • Nice blog ! Your isogonality lemma is really powerful !

    by 554183, Oct 14, 2021, 8:55 AM

  • Post plss....

    by samrocksnature, Apr 11, 2021, 10:12 PM

  • alas,this is ded

    by Hamroldt, Mar 18, 2021, 4:13 PM

  • Thanks for the nice blog.

    by Feridimo, Mar 6, 2020, 4:17 PM

  • I think this might be silly but ... when should we expect to have another post ?? I am very keen to see it :D

    by gamerrk1004, Nov 4, 2019, 4:54 PM

  • Let's all echo what's written in the blog description - Stay Insane / 'Cause it's your labor, will and pain/ That takes you to the top of soda fountain :D

    by Kayak, Oct 2, 2017, 7:18 PM

  • hey utkarsh jee is over now ... continue your elementary blog pleaseeeeeee!

    by kk108, Jun 17, 2017, 11:19 AM

  • Congrats on becoming a contest moderator!

    by Ankoganit, Mar 9, 2017, 5:22 AM

  • INTERSTING BLOG

    by kk108, Feb 19, 2017, 2:04 PM

  • I have no plans for this blog right now....
    No time here people !
    But lets see....
    I may try some combinatorics :P

    by utkarshgupta, Feb 15, 2017, 12:47 PM

  • Thanks for the nice blog!

    by Orkhan-Ashraf_2002, Feb 13, 2017, 6:34 PM

  • Revive it!!!
    Best blog out there, for sure!

    by rmtf1111, Jan 12, 2017, 6:02 PM

48 shouts
Tags
About Owner
  • Posts: 2280
  • Joined: Jan 4, 2013
Blog Stats
  • Blog created: Nov 30, 2013
  • Total entries: 86
  • Total visits: 39606
  • Total comments: 102
Search Blog
a