0
$\begingroup$

In a normal distribution probability curve, it is said that the probability exists for a range of random variables. Probability for a specific point is said to be zero. Can you explain why is this so along with its practical/physical meaning?

$\endgroup$
0
$\begingroup$

There are a number of ways to look at this.

First, intuitively, the "size" of a single point in comparison to the real line is negligible. As an analogy, consider a population of size $10^{500}$. If we choose one element uniformly at random, then that single element occurs with probability $10^{-500} \approx 0$. The real line is "infinitely bigger" than a population of size $10^{500}$.

Formally, this is a consequence of the integral definition. For a continuous random variable $X$ with probability density function $f$, we define $$P(a \leq X \leq b) = \int_a^b f(x)\ dx.$$ If we only care about a single point, then we get $$P(X = a) = P(a \leq X \leq a) = \int_a^a f(x)\ dx = 0.$$ This holds for any continuous random variable, and normally distributed ones in particular.

If you want to get really technical, then this is a consequence of the fact that finite sets have Lebesgue measure zero. In a sense, sets like $\{a\}$ are simply too small to integrate over without getting $0$.

$\endgroup$

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.