0
$\begingroup$

Suppose that we have that $Y=e^{aX}$ where $a$ is a positive scalar. We know that $Y$ follows a logNormal distribution with parameters $0$ and $2$. Then is there a way to derive the distribution of $X$??

From the definition, I think the $log(Y)=aX$ has to follow a normal distribution with mean $0$ and standard error $2$, but this distribution corresponds to $aX$. Thus, I would like to find a distribution of just $X$.

Something that came to my mind right now is that $aX$ (this kind of transformation) results to changes in the variance of the Normal distribution.

$\endgroup$
0
$\begingroup$

If $Y=e^{aX}$ is log-normal with log-mean $0$ and log-variance $2$, then $aX$ is normal with mean $0$ and variance $2$.

Rescale $aX$ by $\frac{1}{a}$, then $X=\frac{1}{a}\times aX$ is normal with mean $\frac{0}{a}=0$ and variance $\frac{2}{a^2}$.

If the parameters you refer to are the log-mean and the log-standard deviation instead of the variance, then the standard deviation of $X$ will be $\frac{2}{|a|}$.

| cite | improve this answer | |
$\endgroup$

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.