Repeat the preceding problem when the seating is random but subject to the constraint that the men and women alternate.

Given that the $C_i$, denote the event that the members of a couple $i$are seated next to each other,$i=1,...,n.$

We have to find$P\left(C_i\right)$

From the combinatorics, we learn that we can permute men on $n!$paths.

Even, we learn that we can permute women on$n!$paths.

Put them around the table in that demand and alternate.

Since the table is rounded, we do not differ in every $2n$seating. Therefore, we have that there exist $\frac{(n!{)}^2}{2n}$modes to seat them. Suppose that the couple $i$has been seated somewhere, but together.

Then, we have to remain $n-1$men and $n-1$women and we can set them on $\left(\right(n-1)!{)}^2$routes since they have to alternate, and also, we do not differ every$2n$seating.

Therefore

$P\left(C_i\right)=\frac{\frac{\left(\right(n-1)!{)}^2}{2n}}{\frac{(n{)}^2}{2n}}=\frac{1}{n^2}$

The probability of $C_i$ is:

Given that $j\ne i$.

We have to find$P\left(C_j\mid C_i\right)$

Assume that pair $i$and $j$sit together.

So, the remaining pairs can place on$\left(\right(n-2)!{)}^2$modes.

Also, as in (a), we have that

$P\left(C_j\mid C_i\right)=\frac{P\left(C_j,C_i\right)}{P\left(C_i\right)}=\frac{\frac{1}{n^2(n-1{)}^2}}{\frac{1}{n^3}}=\frac{1}{(n-1{)}^2}$

Probability of $C_j|C_i$ is,$P\left(C_j\mid C_i\right)=\frac{P\left(C_j,C_i\right)}{P\left(C_i\right)}=\frac{\frac{1}{n^2(n-1{)}^2}}{\frac{1}{n^3}}=\frac{1}{(n-1{)}^2}$

Given that for $n$large, there are no married couples who are seated next to each other.

The probability that some pair sits to each other is $1-P\left(C_i\right)=\frac{n^2-1}{n^2}$. Describe $X$as the random variable that denotes the number of pairs that do not sit to each other.

Utilizing Poisson approximation, we have that $X~$Pois $\left(\frac{n^2-1}{n^2}\right)$. For very large, the required probability is

$P(X=0)=e^{-\lambda }\approx e^{-1}$

since

For $n$large, there are no married couples who are seated next to each other. The approximate probability is

$P(X=0)=e^{-\lambda }\approx e^{-1}$