Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I need a head check on this one. Suppose $\sigma,\mu \in \mathbb{R}$ and $\sigma \neq 0$. Let $X$ be a random variable with density $f_X(x)$. I think that the random variable $Z:= \sigma X + \mu$ has density $$ f_Z(x) = \frac{1}{|\sigma|}f_X\left(\frac{x-\mu}{\sigma} \right). $$

Splitting the cases $\sigma > 0$ and $\sigma<0$ my proof comes down to a simple change of variables. However, I can't find any mention of a formula like this on wikipedia or google. Is it so simple that no one thought to mention it, or am I misunderstanding?

share|improve this question
    
For one thing, you expect that $\int_{\mathbb R} f_Z(x) \, dx = 1$, but if you use the formula for $f_Z(x)$ that you have, you don't get 1. –  echoone Dec 4 '12 at 2:57
    
@echoone And why not? –  Sasha Dec 4 '12 at 3:08
    
@echoone $\int_{\mathbb{R}} f_Z(x)dx = \int_{\mathbb{R}} f_X(x)dx = 1$. Don't forget the Jacobian when you change variables. –  nullUser Dec 4 '12 at 3:09
    
Oops. Yeah, you guys are right. What I am wondering now is how you are dealing with the $\sigma < 0$ case. –  echoone Dec 4 '12 at 3:37
1  
The formula is on wikipedia here en.wikipedia.org/wiki/…. You just have to replace $g$ by the special case of a linear function! –  Learner Dec 4 '12 at 3:44
show 1 more comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.