8
$\begingroup$

Let $f:\mathbb{R^+}\rightarrow \mathbb{R} $ a differentiable function such that $f'(x) = \frac{1}{x}$ and $f(1)=0$ . Show that $f(xy) = f(x) + f(y)$ for all $x,y \in \mathbb{R^+}$

It seems to be the logarithm function but I am unable to prove this statement without the fundamental theorem of calculus, is there any other way to prove this fact?

Thanks for your help

$\endgroup$
10
$\begingroup$

Consider $f(xy)-f(x)$. Differentiating with respect to $x$ yields $yf'(xy)-f'(x)=\frac{y}{xy}-\frac{1}{x}=0$, meaning that $f(xy)-f(x)=C$, where $C$ is a constant. Plug in $x=1$, $f(y)-f(1)=C$ and hence $f(y)=C$. The result follows.

$\endgroup$
  • $\begingroup$ I like your solution, really simple. $\endgroup$ – Usser123456798 Jun 13 '15 at 0:00
11
$\begingroup$

From the definition $f'(x) = \frac{1}{x}$ with $f(0)=1$ we have $$f(x) \equiv \int_1^x \frac{dt}{t}$$

By a simple change of variables $z = ty$ to the integral above we get the result

$$f(x) = \int_y^{xy}\frac{dz}{z} = \int_1^{xy}\frac{dz}{z} - \int_1^y \frac{dz}{z} = f(xy) - f(y)$$


Note that we can also prove the logarithmic property $f(x^r) = r f(x)$ by applying the change of variables $z = t^{1/r}$. Taking it further we can also quite easily show that $f$ has an unique inverse $E(x)$ satisfying $E'(x) = E(x)$ and $E(x+y) = E(x)E(y)$ so $E(x) = e^x$ where $e = E(1)$ is defined by $\int_1^e\frac{dt}{t} = 1$. This gives an alternative definition of the exponential function and of Euler's constant $e$.

$\endgroup$
5
$\begingroup$

Consider $G:\Bbb{R}^+\to\Bbb{R}$ such that $G(x,y)=f(xy)-f(x)-f(y)$.

One has

$${\partial{G(x,y)}\over\partial{x}}={y\over xy}-{1\over x}=0$$ $${\partial{G(x,y)}\over\partial{y}}={x\over xy}-{1\over y}=0$$

Therefore $G$ is constant. $G(x,y)=G(1,1)=0$

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.