0

If X~Expo(1) and S is a random sign (1 or -1 with each p=0.5) and S and are independent. How can I find the PDF of SX if I first should find the CDF of SX? I know that the resulting PDF should be somewhat similar to the Laplace PDF. I have been thinking about starting with something like:

$F(x)=\int_0^x{}S(t)e^{-t}dt$

But don't know if it's correct or how to treat S(t)... Any help would be appreciated!


Is the following correct?

For $x\geq0$, I get:

$F(x)=\frac{1}{2}(P(X\leq{}x)+1)=\frac{1}{2}(1+\int_0^xe^{-t}dt)=1-\frac{1}{2}e^{-x}$

And

For $x\leq0$, I get:

$F(x)=\frac{1}{2}(0+P(X\geq{}-x))=\frac{1}{2}(1-P(X\leq{}-x))=\frac{1}{2}(1-\int_0^{-x}e^{-t}dt)=\frac{1}{2}e^{x}$

Or have I misunderstood something?


A silly question I have is regarding the equation where $\frac{1}{2}$ is introduced. Is correct that we use that since S and X are independent then so are SX and S?

1 Answers1

0

Hint: $$P(SX \le x) = P(SX \le x,S = 1) + P(SX \le x, S = -1) = \frac{1}{2}(P(X \le x) + P(X \ge -x))$$

Now consider the cases $x \ge 0$ and $x \le 0$ separately and remember that the PDF is the derivative of the CDF.

Dominik
  • 19,963