Given a random variable whose density is given by a uniform distribution, can we use the expected value to calculate another expected value?

18 Views Asked by At

If given a random variable X whose density is given by the following uniform distribution,

$$p(x)= \begin{cases} 1 & \ 0 <x < 40 \\ \\ 0 & \text{else} \end{cases}$$

the price of x is x^2. Say we wanted to find the expected cost would we integrate like this

$\int_{0}^{40} x^2 * p(x) \,dx$

and take our final answer as the result of this or like this?

$\int_{0}^{40} x * p(x) \,dx$

and then we would square the latter to get the cost?

They seem very similar yet they yield different results. Can anyone explain what the procedure is here and why?