1
$\begingroup$

If there is a PDF $\text{f}(x)$, and $x$ follows different distributions, for example, Rayleigh, Rice, Nakagami, and Weibull. Then what will be the distribution of $|x|^2$ for these distributions? Please refer me to any book, research paper or any article which helps me with this.

$\endgroup$
4
$\begingroup$

You have some random variable $X$ with density function $f$ and CDF $F$, and you want to know something about the density and/or CDF of the random variable $Y := X^2$. You can do that by considering first the CDF of $Y$.

Let $a$ be a fixed nonnegative number. Then \begin{align*} \mathbb P(Y \leq a) &= \mathbb P(X^2 \leq a) \\ &= \mathbb P(-\sqrt a \leq X \leq \sqrt a) \\ &= F(\sqrt a) - F(-\sqrt a) \end{align*} which gives an explicit formula for the CDF of $Y$ in terms of that of $X$. Since the density is the derivative of the CDF, we can leverage this to get the density function of $Y$, which we'll call $g$: \begin{align*} g(a) &= \frac{\mathrm d}{\mathrm da} \mathbb P(Y \leq a) \\ &= \frac{\textrm d}{\textrm d a} \left[ F(\sqrt a) - F(-\sqrt a) \right] \\ &= f(\sqrt a) \cdot \frac{1}{2 \sqrt a} - f(-\sqrt a) \cdot \frac{-1}{2 \sqrt a} \\ &= \frac{f(\sqrt a) + f(-\sqrt a)}{2 \sqrt a} \\ \end{align*} and in particular if $f$ is symmetric, then we have $$g(a) = \frac{f(\sqrt a)}{\sqrt a}.$$

$\endgroup$
4
$\begingroup$

In general, if $X$ has cdf $F(x) = P(X \leq x)$ and density function $f(x)$, the cdf for $X^2$ will be $P(X^2 \leq x) = P(-\sqrt{x} \leq X \leq \sqrt{x}) = F(\sqrt{x}) - F(-\sqrt{x}).$ Then assuming $F$ is differentiable, the density function for $X^2$ is $$\frac{d}{dx}F(\sqrt{x}) - F(-\sqrt{x}) = f(\sqrt{x}) / (2 \sqrt{x}) + f(-\sqrt{x}) / (2 \sqrt{x}).$$ See also this question and the very detailed answer given there.

$\endgroup$

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.