Proof for maximum order of a sequence of exponential variables is not bounded in probability

154 Views Asked by At

Suppose that X1, X2 ... are infinite sequence of independent exponential random variables with mean u. Let X(n) = max{X1,...,Xn}. Show that the sequence X(n) is not bounded in probability.

I know the cdf of Fx(n) = [(1 - exp(-x/u))]^n not sure how to prove that it is not bounded in probality

1

There are 1 best solutions below

0
On BEST ANSWER

For any $x>0$ $$\mathsf P(Y_n > x)=1-\mathbb P(Y_n\leq x)=1-\left(\mathbb P(X_1\leq x)\right)^n=1-(1-e^{-x/u})^n\to 1$$ as $n\to\infty$ since $(1-e^{-x/u})^n\to 0$.

So $Y_n$ converges to infinity in probability. As well as $Y_n\to\infty$ a.s. since $Y_n$ is nondecreasing sequence. So it is unbounded in probability.