Let $f \colon \mathbb{R} \to \mathbb{R}$ be an exponential function defined by $f(x) = \exp (x)$.

Let $\pi$ be a normal distribution.

How to show that the integration $\int f(x) \pi( \mathrm{d}x)$ of unbounded real-valued function $f$ under Gaussian distribution is always finite?


I'll show you how to do it in the case of the standard normal distribution ${\rm N}(0,1)$, leaving the general case to you.

The integral we wish to evaluate is $$ \int_{x=-\infty}^{x=\infty}dx \ e^{x}. \tfrac 1 {\sqrt{2\pi}} e^{-\frac 1 2 x^2}.$$ The first term is $f(x)$, and the second term is the probability density function for ${\rm N}(0,1)$.

We can combine the two exponentials as $e^{-\frac 1 2 x^2 + x}$, and then complete the square, i.e. write $- \frac 1 2 x^2 + x = -\frac 1 2 (x - 1)^2 + \frac 1 2$. Now, if we substitute $u = x-1$, the integral becomes $$ \int_{u=-\infty}^{u=\infty}du \ \tfrac 1 {\sqrt{2\pi}} e^{-\frac 1 2 u^2 + \frac 1 2} = e^{\frac 1 2} \int_{u=-\infty}^{u=\infty} du \ \tfrac 1 {\sqrt{2\pi}} e^{-\frac 1 2 u^2 } = e^{\frac 1 2}. $$


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.