CDF of a r.v. with a negative power in its PDF

1.8k Views Asked by At

I need to compute a CDF knowing the PDF of a random variable, so nothing fancy.

I find the constant c by stating that:

I find $c = 0.5114$

But when I want to determine the CDF and its bounds, I have a problem

My CDF becomes negative, and for a non-negative function, it is quiet a problem Do you see the problem and could you give me a hint ? Thanks in advance !

1

There are 1 best solutions below

3
On BEST ANSWER

The CDF is an antiderivative of the PDF but generally not the "trivial" one given to us by the elementary calculus rules. To illustrate this, instead of doing your problem, I will do a related problem, namely finding the CDF of a random variable with PDF $f(x)=\begin{cases} 0 & x<0 \\ e^{-x} & x \geq 0 \end{cases}$. Then the CDF is

$$F(x)=\int_{-\infty}^x f(y) dy=\begin{cases} 0 & x<0 \\ \int_0^x e^{-y} dy & x \geq 0 \end{cases}.$$

We compute this integral:

$$\int_0^x e^{-y} dy = -e^{-x}-(-1)=1-e^{-x}.$$

Note the term from the endpoint $0$, which is necessary in order to keep $F$ from being negative, since the "trivial" antiderivative is indeed negative.

Essentially the same thing happens in your problem: there will be a term from the endpoint $1$ which keeps $F$ from being negative.