I know the moment generating function of Gaussian random variable is $$E\{e^{rx}\}=\int^{+\infty}_{-\infty}e^{rx}f(x)dx=e^{mr+r^2\sigma^2/2}$$ where $f(x)$ is PDF of Gaussian R.V., $m$ is mean value and $\sigma$ is standard deviation.

But in my question, some papers refer that that MGF of Gaussian R.V. can be divided into $$\int^0_{-\infty}e^{rx}f(x)dx=\left(e^{mr+r^2\sigma^2/2}\right)\Phi\left(\frac{-m-\sigma^2r}{\sigma}\right)$$ where $\Phi$ is CDF of Gaussian R.V., and I don't understand why that MGF can be divided into above equation.

I think it just have basic principle.

Thank you.


By definition, we have

$$f(x) = \frac{1}{\sqrt{2\pi \sigma^2}} \exp \left(- \frac{(x-m)^2}{2\sigma^2} \right).$$


$$\int_{-\infty}^0 e^{rx} \cdot f(x) \, dx = \frac{1}{\sqrt{2\pi \sigma^2}} \int_{-\infty}^0 \exp \left(rx - \frac{(x-m)^2}{2\sigma^2} \right) \, dx.$$

Now write

$$\begin{align*} rx - \frac{(x-m)^2}{2\sigma^2} &= \frac{1}{2\sigma^2} \bigg[ -(x-(m+\sigma^2 r))^2 + (\sigma^2 r)^2 + 2m\sigma^2 r \bigg]. \end{align*}$$

Setting $\tilde{m} := m+\sigma^2 r$, we see that

$$\int_{-\infty}^0 e^{rx} \cdot f(x) \, dx = \frac{1}{\sqrt{2\pi \sigma^2}} e^{rm +\frac{1}{2} \sigma^2 r^2} \int_{-\infty}^0 \exp \left( -\frac{(x-\tilde{m})^2}{2\sigma^2} \right) \, dx.$$

By a standard substitution this yields

$$\begin{align*} \int_{-\infty}^0 e^{rx} \cdot f(x) \, dx &=e^{rm + \frac{1}{2} \sigma^2 r^2} \int_{-\infty}^{-\frac{\tilde{m}}{\sigma}} \frac{1}{\sqrt{2\pi}} \exp \left( -\frac{z^2}{2} \right) \, dz \\ &=e^{rm + \frac12 \sigma^2 r^2} \cdot \Phi \left( - \frac{\tilde{m}}{\sigma}\right) \\&= e^{rm + \frac{1}{2} \sigma^2 r^2} \cdot \Phi \left( - \frac{m+\sigma^2 r}{\sigma} \right). \end{align*}$$


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.