4
$\begingroup$

I'm reading a definition in DeGroot's book that begins with the statement:

"Let X be a bounded discrete random variable whose p.f. is f."

Then he goes on to define the expectation of X.

However, I cannot find a definition of what is meant by a "bounded" discrete random variable anywhere in the book (after an hour of looking). I do know what a discrete random variable is, but what does the word "bounded" mean in this case?

Thanks.

$\endgroup$
1
  • 1
    $\begingroup$ Bounded just means that there exist a number $M>0$ such that $|X| \le M$ with probability 1. $\endgroup$
    – Stephen Montgomery-Smith
    Jun 2, 2014 at 19:10

1 Answer 1

Reset to default
2
$\begingroup$

Bounded, in this case, means what it means in pretty much every mathematical context - there's a maximum and minimum that the variable never exceeds. If you think about the normal random variable, Z, technically Z could be anything, it's just that as you get farther and farther away from zero, the probability diminishes exponentially.

So, $P(Z > z)$ is never actually equal to zero for any finite z. [EDIT] This means that Z is unbounded.

In a bounded random variable, X, there's some value m and some value M such that:

$P(x < m) = 0$ and $P(x>M) = 0$

$\endgroup$
2
  • $\begingroup$ And this means that the normal random variable Z is unbounded? $\endgroup$
    – David
    Jun 3, 2014 at 16:08
  • $\begingroup$ That's correct. Z is unbounded. Edited my answer to make that clear. Z is also different because it's not discrete, but you already mentioned that you know what a discrete RV is. $\endgroup$
    – Duncan
    Jun 3, 2014 at 16:33

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.