Where does my counting argument fails (Counting of numbers with freshness $k$)?
Doing some recreational math, i was trying to reproduce the combinatorial proof of the identity
$$
n^n=sum_{k=1}^nfrac{n!}{(n-k)!}cdot kcdot n^{n-k-1},,,,(*)
$$
mentioned by OP in the Question Non combinatorial proof of formula for nn
?
To be honest this shouldn't be too difficult:
the number of all sequences with length $n$ out of numbers $1,2,..,n$ is clearly $underbrace{ n cdot n...cdot n}_{n,,times}=n^n$ which establishs the l.h.s. . For the r.h.s we demand that the first $k$ numbers are distinct but that the rest of the sequence can be choosen freely as before (Denote this number by $a_k$). In the end we sum over all $a_k$. I get
$$
a_k=underbrace{ncdot(n-1)...cdot(n-k+1)}_{choose,,k,,numbers\distinct}timesunderbrace{n^{n-k}}_{choose,,n-k,,numbers\freely}\=
frac{n!}{(n-k)!}n^{n-k} ,,,,(**)
$$
But this is not the same then the terms under the summation sign in $(*)$ so what am i doing wrong?
Edit 1:
I think i got it, i have to discount for the sequences with $k'>k$ distinct numbers (they can still fall out of the random part of my product) .Is this correct?
Edit 2:
To finish this, the correct quantity is given by the difference of sequences with $k$ and $k+1$ distinct numbers:
$$
a_{k+1}-a_{k}=-frac{n!}{n^{n-k}}left(frac{1}{(n-k-1)!n}-frac{1}{(n-k)!}right)=-frac{k}{n(n-k)!}frac{n!}{n^{n-k}}
$$
Done!
Edit3:
To avoid algebra we can also reason as follows (credit to @darijgrinberg):
Add to $(**)$ the following condtion: the $k+1$ number is fixed to ne one of the numbers in the first part of the product so that we don't get a sequence with $k+1$ distinct numbers. There are $k$ possiblities. Afterwards we can choose all remaining $n-k-1$ numbers at our wish. This results in a replacement $n^{n-k}rightarrow ktimes n^{n-k-1}$
combinatorics summation
asked 2 days ago
tiredtired
10.5k12043
add a comment |
Doing some recreational math, i was trying to reproduce the combinatorial proof of the identity
$$
n^n=sum_{k=1}^nfrac{n!}{(n-k)!}cdot kcdot n^{n-k-1},,,,(*)
$$
mentioned by OP in the Question Non combinatorial proof of formula for nn
?
To be honest this shouldn't be too difficult:
the number of all sequences with length $n$ out of numbers $1,2,..,n$ is clearly $underbrace{ n cdot n...cdot n}_{n,,times}=n^n$ which establishs the l.h.s. . For the r.h.s we demand that the first $k$ numbers are distinct but that the rest of the sequence can be choosen freely as before (Denote this number by $a_k$). In the end we sum over all $a_k$. I get
$$
a_k=underbrace{ncdot(n-1)...cdot(n-k+1)}_{choose,,k,,numbers\distinct}timesunderbrace{n^{n-k}}_{choose,,n-k,,numbers\freely}\=
frac{n!}{(n-k)!}n^{n-k} ,,,,(**)
$$
But this is not the same then the terms under the summation sign in $(*)$ so what am i doing wrong?
Edit 1:
I think i got it, i have to discount for the sequences with $k'>k$ distinct numbers (they can still fall out of the random part of my product) .Is this correct?
Edit 2:
To finish this, the correct quantity is given by the difference of sequences with $k$ and $k+1$ distinct numbers:
$$
a_{k+1}-a_{k}=-frac{n!}{n^{n-k}}left(frac{1}{(n-k-1)!n}-frac{1}{(n-k)!}right)=-frac{k}{n(n-k)!}frac{n!}{n^{n-k}}
$$
Done!
Edit3:
To avoid algebra we can also reason as follows (credit to @darijgrinberg):
Add to $(**)$ the following condtion: the $k+1$ number is fixed to ne one of the numbers in the first part of the product so that we don't get a sequence with $k+1$ distinct numbers. There are $k$ possiblities. Afterwards we can choose all remaining $n-k-1$ numbers at our wish. This results in a replacement $n^{n-k}rightarrow ktimes n^{n-k-1}$
combinatorics summation
asked 2 days ago
tiredtired
10.5k12043
2
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
1
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago
add a comment |
Doing some recreational math, i was trying to reproduce the combinatorial proof of the identity
$$
n^n=sum_{k=1}^nfrac{n!}{(n-k)!}cdot kcdot n^{n-k-1},,,,(*)
$$
mentioned by OP in the Question Non combinatorial proof of formula for nn
?
To be honest this shouldn't be too difficult:
the number of all sequences with length $n$ out of numbers $1,2,..,n$ is clearly $underbrace{ n cdot n...cdot n}_{n,,times}=n^n$ which establishs the l.h.s. . For the r.h.s we demand that the first $k$ numbers are distinct but that the rest of the sequence can be choosen freely as before (Denote this number by $a_k$). In the end we sum over all $a_k$. I get
$$
a_k=underbrace{ncdot(n-1)...cdot(n-k+1)}_{choose,,k,,numbers\distinct}timesunderbrace{n^{n-k}}_{choose,,n-k,,numbers\freely}\=
frac{n!}{(n-k)!}n^{n-k} ,,,,(**)
$$
But this is not the same then the terms under the summation sign in $(*)$ so what am i doing wrong?
Edit 1:
I think i got it, i have to discount for the sequences with $k'>k$ distinct numbers (they can still fall out of the random part of my product) .Is this correct?
Edit 2:
To finish this, the correct quantity is given by the difference of sequences with $k$ and $k+1$ distinct numbers:
$$
a_{k+1}-a_{k}=-frac{n!}{n^{n-k}}left(frac{1}{(n-k-1)!n}-frac{1}{(n-k)!}right)=-frac{k}{n(n-k)!}frac{n!}{n^{n-k}}
$$
Done!
Edit3:
To avoid algebra we can also reason as follows (credit to @darijgrinberg):
Add to $(**)$ the following condtion: the $k+1$ number is fixed to ne one of the numbers in the first part of the product so that we don't get a sequence with $k+1$ distinct numbers. There are $k$ possiblities. Afterwards we can choose all remaining $n-k-1$ numbers at our wish. This results in a replacement $n^{n-k}rightarrow ktimes n^{n-k-1}$
combinatorics summation
asked 2 days ago
tiredtired
10.5k12043
Doing some recreational math, i was trying to reproduce the combinatorial proof of the identity
$$
n^n=sum_{k=1}^nfrac{n!}{(n-k)!}cdot kcdot n^{n-k-1},,,,(*)
$$
mentioned by OP in the Question Non combinatorial proof of formula for nn
?
To be honest this shouldn't be too difficult:
the number of all sequences with length $n$ out of numbers $1,2,..,n$ is clearly $underbrace{ n cdot n...cdot n}_{n,,times}=n^n$ which establishs the l.h.s. . For the r.h.s we demand that the first $k$ numbers are distinct but that the rest of the sequence can be choosen freely as before (Denote this number by $a_k$). In the end we sum over all $a_k$. I get
$$
a_k=underbrace{ncdot(n-1)...cdot(n-k+1)}_{choose,,k,,numbers\distinct}timesunderbrace{n^{n-k}}_{choose,,n-k,,numbers\freely}\=
frac{n!}{(n-k)!}n^{n-k} ,,,,(**)
$$
But this is not the same then the terms under the summation sign in $(*)$ so what am i doing wrong?
Edit 1:
I think i got it, i have to discount for the sequences with $k'>k$ distinct numbers (they can still fall out of the random part of my product) .Is this correct?
Edit 2:
To finish this, the correct quantity is given by the difference of sequences with $k$ and $k+1$ distinct numbers:
$$
a_{k+1}-a_{k}=-frac{n!}{n^{n-k}}left(frac{1}{(n-k-1)!n}-frac{1}{(n-k)!}right)=-frac{k}{n(n-k)!}frac{n!}{n^{n-k}}
$$
Done!
Edit3:
To avoid algebra we can also reason as follows (credit to @darijgrinberg):
Add to $(**)$ the following condtion: the $k+1$ number is fixed to ne one of the numbers in the first part of the product so that we don't get a sequence with $k+1$ distinct numbers. There are $k$ possiblities. Afterwards we can choose all remaining $n-k-1$ numbers at our wish. This results in a replacement $n^{n-k}rightarrow ktimes n^{n-k-1}$
combinatorics summation
combinatorics summation
asked 2 days ago
tiredtired
10.5k12043
asked 2 days ago
tiredtired
10.5k12043
edited 2 days ago
tired
asked 2 days ago
tiredtired
10.5k12043
asked 2 days ago
tiredtired
10.5k12043
asked 2 days ago
tiredtired
10.5k12043
10.5k12043
2
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
1
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago
add a comment |
2
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
1
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago
2
2
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
1
1
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago
add a comment |
1 Answer
1
active
oldest
votes
I cannot tell if you already figured it out, but in case you haven't:
$$
a_k = underbrace{ncdot (n-1)cdotldotscdot (n-k+1)}_text{choose first $k$ numbers distinct}timesunderbrace{k}_{begin{array}{c}text{choose $(k+1)^{st}$ number equal to }\text{one of the first $k$ numbers}end{array}}times underbrace{n^{n-k-1}}_text{choose remaining numbers freely}
$$
The middle term must be $k$ because you need the "freshness" to be exactly $k$, meaning the first $k$ numbers are distinct, but the first $k+1$ numbers are not. Therefore, the $(k+1)^{st}$ number needs to be equal to one of the previous $k$.
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3062745%2fwhere-does-my-counting-argument-fails-counting-of-numbers-with-freshness-k%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
I cannot tell if you already figured it out, but in case you haven't:
$$
a_k = underbrace{ncdot (n-1)cdotldotscdot (n-k+1)}_text{choose first $k$ numbers distinct}timesunderbrace{k}_{begin{array}{c}text{choose $(k+1)^{st}$ number equal to }\text{one of the first $k$ numbers}end{array}}times underbrace{n^{n-k-1}}_text{choose remaining numbers freely}
$$
The middle term must be $k$ because you need the "freshness" to be exactly $k$, meaning the first $k$ numbers are distinct, but the first $k+1$ numbers are not. Therefore, the $(k+1)^{st}$ number needs to be equal to one of the previous $k$.
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
add a comment |
I cannot tell if you already figured it out, but in case you haven't:
$$
a_k = underbrace{ncdot (n-1)cdotldotscdot (n-k+1)}_text{choose first $k$ numbers distinct}timesunderbrace{k}_{begin{array}{c}text{choose $(k+1)^{st}$ number equal to }\text{one of the first $k$ numbers}end{array}}times underbrace{n^{n-k-1}}_text{choose remaining numbers freely}
$$
The middle term must be $k$ because you need the "freshness" to be exactly $k$, meaning the first $k$ numbers are distinct, but the first $k+1$ numbers are not. Therefore, the $(k+1)^{st}$ number needs to be equal to one of the previous $k$.
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
add a comment |
I cannot tell if you already figured it out, but in case you haven't:
$$
a_k = underbrace{ncdot (n-1)cdotldotscdot (n-k+1)}_text{choose first $k$ numbers distinct}timesunderbrace{k}_{begin{array}{c}text{choose $(k+1)^{st}$ number equal to }\text{one of the first $k$ numbers}end{array}}times underbrace{n^{n-k-1}}_text{choose remaining numbers freely}
$$
The middle term must be $k$ because you need the "freshness" to be exactly $k$, meaning the first $k$ numbers are distinct, but the first $k+1$ numbers are not. Therefore, the $(k+1)^{st}$ number needs to be equal to one of the previous $k$.
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
I cannot tell if you already figured it out, but in case you haven't:
$$
a_k = underbrace{ncdot (n-1)cdotldotscdot (n-k+1)}_text{choose first $k$ numbers distinct}timesunderbrace{k}_{begin{array}{c}text{choose $(k+1)^{st}$ number equal to }\text{one of the first $k$ numbers}end{array}}times underbrace{n^{n-k-1}}_text{choose remaining numbers freely}
$$
The middle term must be $k$ because you need the "freshness" to be exactly $k$, meaning the first $k$ numbers are distinct, but the first $k+1$ numbers are not. Therefore, the $(k+1)^{st}$ number needs to be equal to one of the previous $k$.
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
answered 2 days ago
Mike EarnestMike Earnest
20.6k11950
20.6k11950
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
add a comment |
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
Thanks, this is exactely what i added to my Question a few seconds ago. So now i have two different Proofs relying purely on combinatorics, cool!!! :)
– tired
2 days ago
1
1
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
@tired Indeed! Since you are doing this for recreation, let me mention one other connection. Let $f_{n,k}=binom{n}k kn^{n-k-1}$, then the formula can be written $$n^n=sum_{k=1}^n k! f_{n,k}.$$ This is interesting because $f_{n,k}$ has a combinatorial interpretation; it is the number of "rooted forests" (forest with a root in each component) on a set of size $n$ with $k$ components. This shows that a function can be described as rooted forest together with a permutation $pi$ of the roots. The correspondence is: if $v$ is a not a root, $f(v)$ is its parent; if $v$ is a root, $f(v)=pi(v)$.
– Mike Earnest
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
this is cool indeed. I wish i would have done more of this stuff back at University. Another Thing that i found nice is to divide everything by $n^n$ and use Stirling. We then see that the probability to get sequences with freshness $k$ is given by $k/n$ for sufficently small $k/n$ ...
– tired
2 days ago
1
1
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
@tired That probability distribution is cool, too! Letting $X_n$ be the freshness, then $P(X_n/sqrt{n}>t)to e^{-t^2/2}$ as $ntoinfty$, so $X_n$ has an asymptotically Weibull distribution. Then $E[X_n]sim sqrt{npi/2}$, so in a random function, around $sqrt{npi/2}$ elements will be contained in a cycle, meaning $f^m(x)=x$ for some $m$.
– Mike Earnest
2 days ago
add a comment |
Thanks for contributing an answer to Mathematics Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Some of your past answers have not been well-received, and you're in danger of being blocked from answering.
Please pay close attention to the following guidance:
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3062745%2fwhere-does-my-counting-argument-fails-counting-of-numbers-with-freshness-k%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
2
You don't choose the last $n-k$ numbers freely. You have to make sure that the first of them violates the distinctness condition, i.e., equals one of the preceding $k$ numbers. That's where the $k$ factor will come from.
– darij grinberg
2 days ago
@darijgrinberg thanks: i was already suspecting that. See my edit!
– tired
2 days ago
@darijgrinberg do you have any idea how to see the last Formula (after edit) directly without resorting to annoying Algebra? Thx!
– tired
2 days ago
1
You don't have to subtract anything. Just fix your argument as I suggested in the first comment.
– darij grinberg
2 days ago
apart from the sign, edit 2 is correct.
– G Cab
2 days ago