MHB Determine all real valued differentiable functions f(x)+f(y)=f(xy)

AI Thread Summary
The discussion focuses on finding all real-valued differentiable functions f defined for x > 0 that satisfy the equation f(x) + f(y) = f(xy) for all x, y > 0. By setting x = y = 1, it is established that f(1) = 0. Differentiating the equation with respect to x leads to the conclusion that xf'(x) is constant, allowing for the integration to yield f(x) = c ln x + d. Given that f(1) = 0, it follows that d must equal 0, resulting in the final solution f(x) = c ln x. These functions are the only solutions to the problem.
lfdahl
Gold Member
MHB
Messages
747
Reaction score
0
Determine, with proof, all the real-valued differentiable functions $f$, defined
for real $x > 0$, which satisfy $f(x) + f(y) = f(xy)$ for all $x, y > 0$.
 
Mathematics news on Phys.org
The first thing I would notice is that, taking y= 0, f(x)+ f(0)= f(0) so that f(x)= 0 for all x.
 
because f(xy) = f(x) + f(y) and defined for x, y >0 I see a log function $f(x) =m\log\,x$ satisfies criteria where m is arbitrary constant. m= 0 gives above solution(post #2)
 
lfdahl said:
Determine, with proof, all the real-valued differentiable functions $f$, defined
for real $x > 0$, which satisfy $f(x) + f(y) = f(xy)$ for all $x, y > 0$.
[sp]With $x=y=1$, $f(1) + f(1) = f(1)$, from which $f(1) = 0$.

Fix $y$, and differentiate the equation $f(x) + f(y) = f(xy)$ with respect to $x$: $f'(x) = yf'(xy)$. Therefore $xf'(x) = xyf'(xy)$.

Now let $z = xy$, so that $xf'(x) = zf'(z)$. Since that is true for all positive numbers $x$ and $z$, it follows that $xf'(x)$ is constant, say $xf'(x) = c$. Then $f'(x) = \dfrac cx$, so we can integrate to get $f(x) = c\ln x + d$ (where $d$ is another constant). But $f(1) = 0$, so that $d=0$. Thus $f(x) = c\ln x$, and those are the only solutions.[/sp]
 
kaliprasad said:
because f(xy) = f(x) + f(y) and defined for x, y >0 I see a log function $f(x) =m\log\,x$ satisfies criteria where m is arbitrary constant. m= 0 gives above solution(post #2)
Hi, kaliprasad!
Your intuitive solution is correct indeed! Thanks for participating!

- - - Updated - - -

Opalg said:
[sp]With $x=y=1$, $f(1) + f(1) = f(1)$, from which $f(1) = 0$.

Fix $y$, and differentiate the equation $f(x) + f(y) = f(xy)$ with respect to $x$: $f'(x) = yf'(xy)$. Therefore $xf'(x) = xyf'(xy)$.

Now let $z = xy$, so that $xf'(x) = zf'(z)$. Since that is true for all positive numbers $x$ and $z$, it follows that $xf'(x)$ is constant, say $xf'(x) = c$. Then $f'(x) = \dfrac cx$, so we can integrate to get $f(x) = c\ln x + d$ (where $d$ is another constant). But $f(1) = 0$, so that $d=0$. Thus $f(x) = c\ln x$, and those are the only solutions.[/sp]

Thanks, Opalg! for your participation. Your solution is - of course - correct!(Yes)
 
Suppose ,instead of the usual x,y coordinate system with an I basis vector along the x -axis and a corresponding j basis vector along the y-axis we instead have a different pair of basis vectors ,call them e and f along their respective axes. I have seen that this is an important subject in maths My question is what physical applications does such a model apply to? I am asking here because I have devoted quite a lot of time in the past to understanding convectors and the dual...
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Back
Top