Here is a question about probability density. I am trying to work it out using a different method from the method on the textbook. But I get a different answer unfortunately. Can anyone help me out?

Question: Let $X$ be uniformly distributed random variable in the internal $[ 0, 1]$. Find the probability density of $X^2$?

My trial: $$\begin{align} F_Y(y) & = P(Y\leq y) \\ & = P(X^2\leq y) \\ & = P(X\leq\surd y)\\ & = F_X(\surd y) \\ & = \int_0^{\surd y} t \;\mathrm d t \\ & = 0.5 y\\ \therefore f_Y(y) & =0.5 \end{align}$$ This is actually inspired by https://math.stackexchange.com/questions/...

Solution on the textbook: $$\begin{align} y & = x^2 \\ \mathrm dy & = 2x \;\mathrm dx \\ h(y)\;\mathrm dy & = 1 \mathrm dx \\ h(y) 2x \;\mathrm dx & = \mathrm dx \\ h(y) & = 0.5/x \\ & = 0.5/\surd y \end{align}$$


The method is fine in principle, but the wrong density function for $X$ was used..\ As in your argument, we have, for $0\lt y\lt 1$, $$F_Y(y)=\Pr(X\le \sqrt{Y}).$$ However, if you really want to use an integral, $$F_X(\sqrt{y})=\int_0^{\sqrt{y}} 1\cdot dt=\sqrt{y}.$$ This is because the density function of $X$ on $(0,1)$ is $1$.

Differentiating, we find that $f_Y(y)=\frac{1}{2y^{1/2}}$ for $0\lt y\lt 1$.


My Trial $$\begin{align} F_Y(y) & = P(Y\leq y) \\[1ex] & = P(X^2\leq y) \\[1ex] & = P(X\leq\surd y)\\[1ex] & = F_X(\surd y) \\[1ex] & = \int_0^{\surd y} t \;\mathrm d t & \bigstar \\[1ex] & = 0.5 y\\[2ex] \therefore f_Y(y) & =0.5 \end{align}$$

$\bigstar$ Your error is here. $f_X(t) = 1$ for all $t\in[0;1]$, and $F_X(t) = t$; so $F_X(\surd y)=\surd y$.

$$\begin{align} F_Y(y) & = F_X(\surd y) \\[1ex] & = \int_0^{\surd y} 1 \;\mathrm d t & \bigstar \\[1ex] & = \surd y\\[2ex] \therefore f_Y(y) & =1 / (2\surd y) \end{align}$$

Addendum. The textbook's solution demonstrates that if you have an invertable map of $X\to Y$ and a pdf for $X$ and wish to find the pdf for $Y$, then you don't have to integrate then derive again, you just need to apply a change of variable transformation (one derivation).

$$Y=g(X) \implies f_Y(y) = f_X(g^{-1}(y)) \left\lvert\frac{\mathrm d g^{-1}(y)}{\mathrm d y}\right\rvert$$

This will often be a much easier approach, particularly if the initial distribution is something other than uniform.   Learn it and love it.

$$\begin{align}Y=X^2, f_X(x)=\mathbf 1_{(0;1)}(x) \implies f_Y(y) & = f_X(\surd y) \left\lvert\frac{\mathrm d \surd y}{\mathrm d y}\right\rvert \\ & = \frac{1}{2\surd y}\;\mathbf 1_{(0;1)}(y) \end{align}$$


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.