Probability Density Function and IID Random Variables that go to infinite

69 Views Asked by At

Im new to Statistic, please just dont blast me if you think my question is stupid :(

Image i take n random variables (IID and continuos) . Is there a theorem that assures that if n goes to infinite, the distribution of the values of these random variables follow exactly the probability density function ?

1

There are 1 best solutions below

0
On

Almost. You have to translate your statement into probabilities and limits

For example if you have a random variable $X$ with a cumulative distribution function $F(x)$ then the probability $p_{x_0,\delta}$ that the random variable takes a value in the interval $(x_0-\delta , x_0+\delta]$ around a given $x_0$ is $p_{x_0,\delta}=\Pr(x_0-\delta \lt X_n \le x_0+\delta) = F(x_0+\delta)-F(x_0-\delta)$

Now let $I_{n}$ be the indicator random variable taking the value $1$ when $x_0-\delta \lt X_n \le x_0+\delta$ and the value $0$ otherwise. Then the laws of large numbers say that $\frac1n \sum_{j=1}^n I_j$ converges in probability and almost surely to $p_{x_0,\delta}$ as $n$ increases

If $X$ has a density function $f(x)$ then $p_{x_0,\delta}=\int_{x_0-\delta}^{x_0+\delta} f(x)\, dx$. If the density function is continuous at $x_0$ then it is the derivative of the cumulative distribution function and $f(x_0) =\lim_{\delta\to 0} \frac{p_{x_0,\delta}}{2\delta}$