14
$\begingroup$

Let $a, n$ be positive integers with $a < n$. A revolver with $n$ chambers is loaded with $a$ bullets, where the distribution is uniform among all $\binom{n}{a}$ possible choices of $a$ objects filling $n$ slots.

A lone player is playing Russian Roulette using the loaded gun. In this game the gun is spun, which we conceptualise as a chamber being picked uniformly at random. Subsequently, the player fires at his own head without spinning the gun again - which we conceptualise as going through the chambers deterministically in sequence.

In this game, the player intends to shoot himself until he dies. We assume that there is no chance of surviving a headshot - the player dies the moment a non-empty chamber is encountered.

We denote by $X^{(a, n)}_t$ the survival process of a player using the loaded gun, where at each time $t$, $X^{(a, n)}_t$ takes value $1$ if the player is still alive after the $t$’th round, and $0$ otherwise.

Write $S^{(a, n)} := \sum_{t = 1}^{\infty} X^{(a, n)}_t$ for the survival time of the player.

Question: Is it true that the mean survival time $\mathbb E[S]$ is not monotone decreasing in the ratio $\frac{a}{n}$ of bullets to chambers?

That is, does there exist $(a, n), (a^{*}, n^{*})$ with $\frac{a^{*}}{n^{*}} > \frac{a}{n}$, but

$$\mathbb E[S^{(a^*, n^*)}] > \mathbb E[S^{(a, n)}]?$$

Bonus points if it can be shown that $n$ can be taken arbitrarily large.

$\endgroup$

1 Answer 1

14
$\begingroup$

Let $S$ be the survival time. Then $$P(S\ge s)=\binom{n-s}a\Big/\binom na$$ for $s=0,1,\dots$. So, $$ES=-1+\sum_{s=0}^\infty P(S\ge s)=\frac{n+1}{a+1}-1.$$

So, $n=3$, $a=1$, $n^*=8$, $a^*=3$ will do.


More generally, any integers $n$, $a$, $m:=n^*$, $b:=a^*$ such that $n\ge2$, $1\le a<n$, $m>n$, and $$\frac{ma}n<b<\frac{ma+m+a-n}{n+1} \tag{10}\label{10} $$ will do. Note that $$\frac{ma+m+a-n}{n+1}-\frac{ma}n=\frac{(m-n) (n-a)}{n (n+1)}>1$$ if $m$ is large enough, so that then there is an integer $b$ satisfying \eqref{10}. So, indeed, $n,a,m,b$ can be taken arbitrarily large.

$\endgroup$
7
  • $\begingroup$ Ah, and the numerator comes from the following procedure - fix a slot that we determinsticslly start at and offload all of the randomness onto the bullet filling procedure. Then we will survive $s$ rounds iff the first $s$ slots are empty, leading to $a$ objects filling $n-s$ holes. $\endgroup$
    – Nate River
    Commented Nov 6 at 3:32
  • $\begingroup$ Also, haha I made a mistake with the original notation, if you don’t mind I’ll change yours a bit to fit the corrected notation. $\endgroup$
    – Nate River
    Commented Nov 6 at 3:33
  • $\begingroup$ @NateRiver : That made me do a lot of retyping. :-) $\endgroup$ Commented Nov 6 at 3:45
  • $\begingroup$ what is $\binom{a}{n}$ for $a<n$? $\endgroup$
    – sds
    Commented Nov 6 at 17:31
  • 3
    $\begingroup$ @sds : Thank you for your comment. After this answer was posted, the OP interchanged the roles of $a$ and $n$. I tried to edit the answer accordingly, but apparently missed this. Now this typo should be fixed. $\endgroup$ Commented Nov 6 at 17:44

You must log in to answer this question.

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