Bivariate tranformation?

Question

If two random variable (independent) is given say $X$ which is real and distributed $~N(0,1)$ and a discrete random variable $\alpha$ that takes +1 or -1 with probability half each. A transformation is given as $Y=\alpha X$. I need to find the joint distribution between $X$ and $Y$.

Is it through the bivariate transformation considering one tranformation as $U=\alpha X$ and another one as V=X and I can find their joint distribution or any easy intuitive way to figure out this?

I am following this : https://onlinecourses.science.psu.edu/stat414/node/129

score 0 · Accepted Answer · answered Jan 11 '17 at 07:48

0

\begin{align} P(X \leq x, Y \leq y) &=P(\alpha=1)P(X \leq x, \alpha X \leq y| \alpha =1)+ P(\alpha=-1)P(X \leq x, \alpha X \leq y| \alpha =-1)\\ &=\frac12 P(X \leq x, X \leq y)+ \frac12P(X \leq x, -X \leq y) \\ &=\frac12 P(X \leq \min(x,y))+ \frac12 P(-y \leq X \leq x) \end{align}

answered Jan 11 '17 at 07:48

Siong Thye Goh

149,520
20
88
149

thanks!, by the way y will be always less than or equal o x, right? so min(x,y)=y, isnt it and the joint distribution is Gaussian? – Nikhil cherian kurian Jan 11 '17 at 08:36
$Y \leq X$ but $x$ and $y$ are numbers that we want to evaluate the cdf, so $y$ need not be smaller than $x$. – Siong Thye Goh Jan 11 '17 at 08:39
Thanks understood, Sorry to ask you, can you suggest a good reference on probabiity random variable, I am a beginner in this course. – Nikhil cherian kurian Jan 11 '17 at 08:42
I used to browsed through Sheldon Ross "A first course in probability" , Bersekas "Introduction to probability", notes by David Gamarnik (MIT OCW fundamentals of probability). – Siong Thye Goh Jan 11 '17 at 08:49

Bivariate tranformation?

1 Answers1