1
$\begingroup$

Has anyone seen this notation before? What does it mean?

$\int_{0}^{\infty} f(x) G(x) dx$

$f(x)$ is a density and $G(x)$ is a cumulative distribution function

$\endgroup$
  • $\begingroup$ Can you tell us a little more? In which context did you see this? is $G(x)$ the cdf corresponding to the density $f(x)$, or is it some other cdf? $\endgroup$ – kjetil b halvorsen Feb 7 '16 at 19:12
  • $\begingroup$ I have something like $f(x)=pr(I'=k, x \leq M'<x+dx)$ and $G(x)=pr(M \leq x)$ $\endgroup$ – Nick Feb 7 '16 at 19:17
  • 1
    $\begingroup$ I interpreted $f(x)$ and $G(x)$ as density and cdf. I am not sure if I am correct. What does it mean? Any reference is appreciated. $\endgroup$ – Nick Feb 7 '16 at 19:24
1
$\begingroup$

Suppose that $X$ and $Y$ are independent continuous random variables with pdfs $f$ and $g$ respectively, and that $G$ is the CDF of $Y$. Suppose also that $X \geq 0$.

Then, \begin{align} P\{Y \leq X\} & = \int_{-\infty}^\infty \int_{-\infty}^x f(x)g(y)\, \mathrm dy\,\mathrm dx\\ &= \int_{-\infty}^\infty f(x) \left[\int_{-\infty}^x g(y)\, \mathrm dy\right]\,\mathrm dx\\ &= \int_{-\infty}^\infty f(x) G(x)\,\mathrm dx\\ &= \int_0^\infty f(x) G(x)\,\mathrm dx &\scriptstyle{\text{since}~ f(x) = 0~\text{when}~ x < 0} \end{align} Note that it is not necessary to assume that $Y$ is positive (having $X \geq 0$ suffices), and that we get the expected value of $P\{Y \leq X\}$ (as in kjetil b halvorsen's answer) only in the sense that $P\{Y \leq X\}$ is a constant whose expected value is just that constant.

If $X$ can take on both positive and negative values, then $$\int_0^\infty f(x) G(x)\,\mathrm dx = P\left(\{Y \leq X\}\cap \{X \geq 0\}\right).$$

$\endgroup$
1
$\begingroup$

I assume both the density $f(x)$ and the cdf $G(x)$ refers to random variables on the interval $(0, \infty)$. Let $X$ be a random variable with the density $f$ and $Y$ be a random variable with the cdf $G $. Then $$ \DeclareMathOperator{\E}{\mathbb{E}} \DeclareMathOperator{\P}{\mathbb{P}} \int_0^\infty f(x) G(x) \; dx = \E_X G(X) = \E_X \P (Y \le X) $$ If you could tell us some more about the context where you saw this, we might tell you why it could be useful.

$\endgroup$
  • $\begingroup$ Does your comment apply in the situation where $f(x)dx=pr(I′=k,x \leq M′<x+dx)$ and $G(x)=pr(M \leq x)$? $\endgroup$ – Nick Feb 7 '16 at 19:31
  • $\begingroup$ I show this in the book an introduction to probability and applications. I do not have anything more. $\endgroup$ – Nick Feb 7 '16 at 19:52

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.