Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
80 changes: 17 additions & 63 deletions src/content/questions/comp2804/2022-winter-final/15/solution.md
Original file line number Diff line number Diff line change
@@ -1,63 +1,17 @@
Warning: I don't know but I can try

<ul>
<li> Let S be the set of all possible permutations <br/>
$ |S| = 20! $
<li> Let's determine $A$ <br/>
We could pick 1 and any of the 19 larger integers and the other 18 numbers have 18! posible permutations <br/>
We could pick 2 and any of the 18 larger integers and the other 18 numbers have 18! posible permutations <br/>
\ldots <br/>
We could pick 19 and 20 and the other 18 numbers have 18! posible permutations <br/>
$ |A| = 19 \cdot 18! + 18 \cdot 18! + \ldots + 1 \cdot 18! $ <br/>
$ |A| = 18! \cdot \frac{(19)(20)}{2}$ <br/>
$ |A| = 18! \cdot (19)(10) $ <br/>
$ |A| = 18! \cdot 190 $
$ |A| = 10 \cdot 19! $
$ \Pr(A) = \frac{|A|}{|S|} $ <br/>
$ \Pr(A) = \frac{10 \cdot 19!}{20!} $ <br/>
$ \Pr(A) = \frac{10}{20} $ <br/>
$ \Pr(A) = \frac{1}{2} $
<li> Let's determine $\Pr(B)$ <br/>
We could pick 1 and any of the 19 larger integers and the other 18 numbers have 18! posible permutations <br/>
We could pick 2 and any of the 18 larger integers and the other 18 numbers have 18! posible permutations <br/>
\ldots <br/>
We could pick 19 and 20 and the other 18 numbers have 18! posible permutations <br/>
$ |A| = 19 \cdot 18! + 18 \cdot 18! + \ldots + 1 \cdot 18! $ <br/>
$ |A| = 18! \cdot \frac{(19)(20)}{2}$ <br/>
$ |A| = 18! \cdot (19)(10) $ <br/>
$ |A| = 18! \cdot 190 $
$ |A| = 10 \cdot 19! $
$ \Pr(A) = \frac{|A|}{|S|} $ <br/>
$ \Pr(A) = \frac{10 \cdot 19!}{20!} $ <br/>
$ \Pr(A) = \frac{10}{20} $ <br/>
$ \Pr(A) = \frac{1}{2} $
<li> Let's determine $\Pr(A \cap B)$ <br/>
Well, let's go I guess!
<ul>
<li> $\pi_{12}$ could be 1 and $\pi_{11}$ would have to be any of the 19 values smaller than it and $\pi_{10}$ would have to be any value larger than $\pi_{11}$. The remaining 17 values have 17! permutations
<li> $\pi_{12}$ could be 2 and $\pi_{11}$ would have to be any of the 18 values smaller than it and $\pi_{10}$ would have to be any value larger than $\pi_{11}$. The remaining 17 values have 17! permutations
<li> \ldots
<li> $\pi_{12}$ could be 19 and $\pi_{11}$ would have to be 20 and $\pi_{10}$ would have to be any value larger than $\pi_{11}$. The remaining 17 values have 17! permutations
</ul>
Idk if there's a way to express it but we can try <br/>
$ \sum_{i=1}^{18} ( \sum_{j=i+1}^{19} (\sum_{k=j+1}^{20} 1 \cdot 17!) ) $<br/>
$ 17! \cdot \sum_{i=1}^{18} ( \sum_{j=i+1}^{19} (\sum_{k=j+1}^{20} 1) ) $<br/>
$ 17! \cdot \sum_{i=1}^{20} ( \sum_{j=1}^{i-1} (\sum_{k=1}^{j-1} 1) ) $<br/>
$ 17! \cdot \sum_{i=1}^{20} ( \sum_{j=1}^{i-1} ( j - 1 ) ) $<br/>
$ 17! \cdot \sum_{i=1}^{20} ( \frac{(i-1)(i-2)}{2}) $<br/>
$ 17! \cdot \frac{1}{2} \cdot \sum_{i=1}^{20} ( (i-1)(i-2) ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot \sum_{i=1}^{20} ( i^2 - 3i + 2 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( \sum_{i=1}^{20} i^2 - \sum_{i=1}^{20} 3i + \sum_{i=1}^{20} 2 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( \frac{20 \cdot 21 \cdot 41}{6} - 3 \sum_{i=1}^{20} i + 2 \sum_{i=1}^{20} 1 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( \frac{20 \cdot 21 \cdot 41}{6} - 3 \cdot \frac{20 \cdot 21}{2} + 2 \cdot 20 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( \frac{20 \cdot 7 \cdot 41}{2} - 3 \cdot \frac{10 \cdot 21}{1} + 40 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( \frac{10 \cdot 7 \cdot 41}{1} - 3 \cdot 10 \cdot 21 + 40 ) $<br/>
$ 17! \cdot \frac{1}{2} \cdot ( 10 \cdot 7 \cdot 41 - 3 \cdot 10 \cdot 21 + 40 ) $<br/>
$ 17! \cdot ( 5 \cdot 7 \cdot 41 - 3 \cdot 5 \cdot 21 + 20 ) $<br/>
$ 17! \cdot ( 5 \cdot 7 \cdot 41 - 3 \cdot 5 \cdot 21 + 20 ) $<br/>
$ 17! \cdot ( 1435 - 315 + 420 ) $<br/>
$ 17! \cdot ( 1540 ) $<br/>
ngl, the value is actually wrong. Basically the probability of A intersect B is smaller than the product of the 2 probabilities
</ul>

There's no way this is the actual way to do the question, right? This is bullshit. I am LITERALLY losing my sanity
Here the thing is easy but tricky:

For simplicity lets take a1 a2 and a3 are our permutations of set {1,2,3}

so Pr(A=a1>a2) = 1/2 because swapping the values of the variables a1 and a2 will make the sign turn to > and violating the constraints. so we can swap in 2! ways and for all ways the signs will change except 1 way. ie : a1 > a2.

under same logic we can say for Pr(B=a2>a3) = 1/2.

Similarly for Pr(A intersect B ) = 1/3! = 1/6. because under only 1 condition a1>a2>a3 and swapping there values will lead to something else compared to the condition we want.

so we can see Pr(A intersect B) < Pr(A) * Pr(B)

what for the big numbers now ? like n = 100 or 200 instead of 3?

actually n doesnt put an affect here because permutation is something know as a unform distributions.

So irrespective of the value of n for a subset of k it will always be 1/k! as the probability to get a specific sign sequence.