Re: S_n permutation group
- From: "bargiax" <bargiax@xxxxxxxxxxx>
- Date: Wed, 7 Jun 2006 03:22:04 +0200
"Chip Eastham" <hardmath@xxxxxxxxx> ha scritto nel messaggio
news:1149638887.260891.172930@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
bargiax wrote:
Hello, how many permutations a in S_n are there such that a^2=1 ?
My answer is: there are Bin(n,2)+( Bin(n,2)*Bin(n-2,2) )/2 of such
permutations. Is this result correct?
Thank you.
It doesn't appear to give a correct answer when n = 3.
There are four permutations such that squaring gives
the identity permutation, three transpositions and the
identity itself, when n = 3.
If Bin(n,2) means the binomial coefficient, then Bin(3,2)
should be 3. Conventionally Bin(1,2) is zero, ie. no
ways to choose 2 out of 1 thing. With that interpretation
your formula gives 3 rather than 4.
Note that a permutation satisfies a^2 = identity if and
only if it can be written as a product of disjoint cycles
of length no greater than two.
Following that line of reasoning, one easily sees that
for n = 4, the solutions are (a) the identity, (b) six
different transpositions, and (c) three different pairs
of (disjoint) transpositions, for a total of 10 solutions.
Your formula appears to give an answer of 9.
Thank you Chip, perhaps I've found my mistake: my formula didn't mention the
obvious identical permutation. Indeed we know that a)(i,j)=(j,i) and that b)
(i,j)(k,l)=(k,l)(i,j). So the number of different transpositions of the kind
(a) should be Bin(n,2), the rest should be about (b). Unfortunately I did
not include the identity.
.
- Follow-Ups:
- Re: S_n permutation group
- From: Chip Eastham
- Re: S_n permutation group
- References:
- S_n permutation group
- From: bargiax
- Re: S_n permutation group
- From: Chip Eastham
- S_n permutation group
- Prev by Date: Re: JSH: SF: Finally, surrogate factoring
- Next by Date: Re: Another silly question
- Previous by thread: Re: S_n permutation group
- Next by thread: Re: S_n permutation group
- Index(es):
Relevant Pages
|