1
$\begingroup$

For a function such as $f(x) = e^{x^{2}}$, how do I go about finding an anti-derivative for this that passes through a given point? Graphically, there should be no function that has derivative F'(x) = $e^{x^{2}}$, may I have some hints as to how I am able to tackle such a problem?

$\endgroup$
  • 1
    $\begingroup$ "Graphically, there should be no function that has derivative $F'(x)=e^{x^2}$" - Why not? I see no graphical issue with such a function, but if you elaborate on that point, it might help us understand what you're thinking and give a better answer. $\endgroup$ – Milo Brandt Jan 18 '16 at 6:22
  • $\begingroup$ Well, I was thinking of integrating over all values to find an indefinite integral, but after refreshing the definition from an answer below. I believe there is now if we set the bounds right as the area is finite. I guess what I want to do is a method to find a function whose derivative is $e^x^2$. $\endgroup$ – Mid Jan 18 '16 at 6:41
  • $\begingroup$ There is a function $F(x)$ such that $F'(x)=e^{x^2}$. It is given by $F(x)=\frac{1}{2} \sqrt{\pi }\, \text{erfi}(x)+C$ where appears the imaginary error function. $\endgroup$ – Claude Leibovici Jan 18 '16 at 8:32
1
$\begingroup$

Since $f$ is continuous on $\mathbb{R}$, it is Riemann integrable on $[c,x]$ for any $c \leq x$. By the fundamental theorem of calculus, $f: \mathbb{R} \to \mathbb{R}$ has an antiderivative $F: \mathbb{R} \to \mathbb{R}$ $$F(x) = \int_c^x f(t) \, dt$$ You can view this as a definition.

$\endgroup$
  • $\begingroup$ Oh does this involve F(X) - F(C) =$\int_{a}^{x} e^{t^2} dt$. So with a specific point (a,b) we can have F(a) = b so F(X) - b = $\int_{a}^{x} e^{t^2} dt$. But then I am just stuck with the original integral + a constant b? $\endgroup$ – Mid Jan 18 '16 at 6:32
  • $\begingroup$ It does answer the question, but a bit too simple? $\endgroup$ – Mid Jan 18 '16 at 6:55
  • $\begingroup$ I think so. Note that there are more than one antiderivatives. In my answer, the antiderivative is the one which passes through $(c,0)$. You can add a constant $b$ to the right hand side to make it passes through $(c,b)$. $\endgroup$ – Empiricist Jan 18 '16 at 6:58
  • $\begingroup$ Ah, so it would just be $F(X) = \int_{a}^{x} e^{t^2} dt + b$ would be a proper solution for a function with derivative $e^{t^2}$ that passes through point (a,b)? $\endgroup$ – Mid Jan 18 '16 at 7:03
  • $\begingroup$ Yes, I think so. :) $\endgroup$ – Empiricist Jan 18 '16 at 8:53

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.