0
$\begingroup$

Suppose there is a function defined as $$f(x)= \frac{d}{dx}g(x)$$ similar to how $1/x$ is $ \frac{d}{dx}log(x)$ or $ \frac{1}{1+x^2}$ is $ \frac{d}{dx}\arctan(x)$ or something along those lines. Is there a general process (probably using the inverse function theorem) to find $f^{-1}(x)$?

The reason I ask is because there are more difficult functions that may end up being defined this way that don't have immediately apparent relationships. You could for instance define $e^{-x^{2}}$ as $ \frac{d}{dx}erf(x)$.

$\endgroup$
  • 3
    $\begingroup$ @nikola That's not it. Take $g(x) = x^2$, for instance. Then $f(x) = 2x$ and $f^{-1}(x) = \frac12x$. That's not an antiderivative of $g$. Also, there is no need to be condecending. This site is for asking questions. The entire point is to help people learn. No matter the age or level of the asker. $\endgroup$ – Arthur Oct 1 '18 at 6:38
  • $\begingroup$ Sorry, I am blind.... $\endgroup$ – nikola Oct 1 '18 at 6:46
1
$\begingroup$

I will assume everything is invertible and smooth, at least locally. The following holds for every $a$ and $b$. $$ \int_{f(a)}^{f(b)} f^{-1}(x)dx=bf(b)-af(a)-\int_a^b f(x)dx \\ =(bf(b)-g(b))-(af(a)-g(a)) $$ This can be easily seen by regarding the integration as the area of the region below the graph. You may also say that from the above equation, the antiderivative of $f^{-1}(x)$ is $$ \int f^{-1} (x)dx = xf^{-1}(x)-g\circ f^{-1}(x) $$ You can check the equality by differentiating both sides.

$\endgroup$

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.