1
$\begingroup$

In some probability book I've come across this notation: $E(X)=\int_{-\infty}^{\infty}x d F(x)$, and it's very confusing, when I see other books defining the same concept as: $E(X)=\int_{-\infty}^{\infty}xf(x) dx$

I wonder... what does that kind of notation mean?

$\endgroup$
1
$\begingroup$

In this case, $f(x)$ would be the probability density function and $F(x)$ would be the cumulative distribution function. They are related by $$f(x)=\frac{dF(x)}{dx}\ ,$$ so $dF(x)=f(x)\,dx$ and the two integrals are equivalent.

In practice, to evaluate the first integral you would probably begin by transforming it to the second anyway.

$\endgroup$
  • $\begingroup$ Yes, I knew $f(x)$. But I hadn't ever seen the notation $dF(x)$ in an integral just like that. It makes sense like that. $\endgroup$ – David Molano Mar 2 '15 at 1:50
  • $\begingroup$ One should also mention that there are cases where $F$ is not differentiable. Sometimes, it is then possible to interpret $\int_{-\infty}^\infty x\mathrm{d}F(x)$ as a Riemann–Stieltjes integral. (A common case would be that $F$ is piecewise differentiable, then the integral can be evaluated by considering the pieces individually.) $\endgroup$ – Eike Schulte Mar 2 '15 at 8:54

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.