Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have a standard Poisson process $N(t)$ with arrival rate $\lambda$ per second. Each time an arrival occurs, a fair coin is flipped and a counter is increased or decreased by $1$ depending on the result. If we let $M(t)$ indicate the counter value at time $t$ and let $M(0) = 0$ (just like $N(0) = 0)$, at time $t$, if we knew that $N(t) = n$, then the value of the counter $M(t)$ could be understood as

$$ \# \text{(successes)} - \# \text{(failures)}$$

in $n$ Bernoulli trials with success probability $\frac{1}{2}$ . Using this logic, i need to find the conditional PMF $$\mathrm{Pr}[M(t) = m\mid N(t) = n]$$ and the expression for the PMF of $M(t)$.

I am having trouble on how to approach the problem and how # successes - # failures could be understood in finding the PMF. Any help would be greatly appreciated. Thanks!

share|improve this question
add comment

1 Answer

The probability for $k$ successes in $n$ Bernoulli trials with success probability $\frac12$ is $\binom nk2^{-n}$. The resulting value of $m$ is $k-(n-k)=2k-n$. Thus the conditional probability mass function is

$$ \def\pr{\operatorname{Pr}}\pr[M(t)=m\mid N(t)=n]=\begin{cases}\binom n{(n+m)/2}2^{-n}&n+m\text{ even,}\\\\0&\text{otherwise.}\end{cases} $$

Then by the law of total probability,

$$ \pr[M(t)=m]=\sum_{n=0}^\infty\pr[M(t)=m\mid N(t)=n]\pr[N(t)=n]\;, $$

and substituting $\pr[N(t)=n]=\mathrm e^{-\lambda t}(\lambda t)^n/n!$ yields an expression for $\pr[M(t)=m]$ as a sum over the values of $n$ with the same parity as $m$.

share|improve this answer
    
Great thanks! it makes much more sense now. What I'm having trouble finding now is the expected value of $M(t)$. –  αδζδζαδζασεβ23τ254634ω5234ησςε Dec 14 '12 at 8:21
1  
@Janson If your aim is to compute $E(M(t))$, (1) you should have said so in the question, and (2) none of the above is necessary. –  Did Dec 15 '12 at 10:45
    
@did so how would you do it then? –  αδζδζαδζασεβ23τ254634ω5234ησςε Dec 15 '12 at 21:13
1  
Since the probability of each event being a failure or a success is the same, E(M(t)) = 0. –  Did Dec 16 '12 at 7:32
    
@Janson Why the change of name? –  Did Dec 16 '12 at 8:34
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.