0
$\begingroup$

Question

Let $X_1 , X_2, \cdots , X_n$ be independent and identically distributed with $P(X_i =1)=P(X_i=-1)=p$ and $P(X_i=0)=1-2p$ for all $i=1,2,\cdots,n$. Define $$ a_n=P(\prod_{i=1}^{n}X_i=1), b_n=P(\prod_{i=1}^{n}X_i=-1), c_n=P(\prod_{i=1}^{n}X_i=0)$$

Find $\lim_{n\to\infty} a_n,\lim_{n\to\infty} b_n,\lim_{n\to\infty} c_n$.

My Approach

Since $(\prod_{i=1}^n X_i=1)=1$ hence $a_n=P(\prod_{i=1}^{n}X_i=1)=P(X=1)=p$ and so $\lim_{n\to\infty}a_n=p$.

Similarly for others. Is this approach correct?

$\endgroup$
8
  • $\begingroup$ It is more likely you will get an answer if you show you have done some work or please briefly explain what concept (s) you are having difficulty understanding. $\endgroup$ Commented Jun 13, 2018 at 6:07
  • $\begingroup$ @TonyHellmuth Updated! Sorry! $\endgroup$ Commented Jun 13, 2018 at 6:07
  • $\begingroup$ Also do you have any information to believe that the limits have a non-alternating value? $\endgroup$ Commented Jun 13, 2018 at 6:09
  • $\begingroup$ @TonyHellmuth It's a part of a multiple choice question. Should I post the other options as well? $\endgroup$ Commented Jun 13, 2018 at 6:10
  • 1
    $\begingroup$ Great observation - that is the starting point to attack this question. Consider the case where all of them are non-zero, which help you to compute $c_n$. Next you may need to consider some odd/even issue for $a_n, b_n$. $\endgroup$
    – BGM
    Commented Jun 13, 2018 at 7:41

1 Answer 1

1
$\begingroup$

Since $1-2p$ is a probability, we can assume that $p \leq 1/2$. If $p<1/2$, then $c_n=1-\text{Pr[No $X_i$ is zero]=}1-(2p)^n$. Thus the third limit is 1. The sum of the three quantities is 1, so the first two limits are zero.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .