Increasing and Decreasing Subsequences
Richard P. Stanley
M.I.T.
Increasing and Decreasing Subsequences – p. 1
Definitions
318496725
(i.s)
3 1 8 4 9 6 7 2 5 (d.s)
is(w) = |longest i.s.| = 4
ds(w) = |longest d.s.| = 3
Increasing and Decreasing Subsequences – p. 2
Application: airplane boarding
Naive model: passengers board in order
w = a1 a2 · · · an for seats 1, 2, . . . , n. Each
passenger takes one time unit to be seated after
arriving at his seat.
Increasing and Decreasing Subsequences – p. 3
Boarding process
253614
6
5
4
3
2
1
Increasing and Decreasing Subsequences – p. 4
Boarding process
253614
6
5
4
3
2
1
2536 14
Increasing and Decreasing Subsequences – p. 4
Boarding process
253614
6
5
4
3
2
1
2536 14
5
364
2
1
Increasing and Decreasing Subsequences – p. 4
Boarding process
253614
6
5
4
3
2
1
2536 14
5
6
364
2
1
2
1
4
5
3
Increasing and Decreasing Subsequences – p. 4
Results
Easy: Total waiting time = is(w).
Bachmat, et al.: more sophisticated model.
Increasing and Decreasing Subsequences – p. 5
Results
Easy: Total waiting time = is(w).
Bachmat, et al.: more sophisticated model.
Two conclusions:
Usual system (back-to-front) not much better
than random.
Better: first board window seats, then center,
then aisle.
Increasing and Decreasing Subsequences – p. 5
Partitions
partition λ ⊢ n: λ = (λ1 , λ2 , . . . )
λ1 ≥ λ2 ≥ · · · ≥ 0
X
λi = n
Increasing and Decreasing Subsequences – p. 6
Young diagrams
(Young) diagram of λ = (4, 4, 3, 1):
Increasing and Decreasing Subsequences – p. 7
Conjugate partitions
λ′ = (4, 3, 3, 2), the conjugate partition to
λ = (4, 4, 3, 2)
λ
λ’
Increasing and Decreasing Subsequences – p. 8
Standard Young tableau
standard Young tableau (SYT) of shape λ ⊢ n,
e.g., λ = (4, 4, 3, 1):
<
<
1 2 7 10
3 5 8 12
4
9
6 11
Increasing and Decreasing Subsequences – p. 9
f
λ
f λ = # of SYT of shape λ
E.g., f (3,2) = 5:
123
45
124
35
125
34
134
25
135
24
Increasing and Decreasing Subsequences – p. 10
f
λ
f λ = # of SYT of shape λ
E.g., f (3,2) = 5:
123
45
124
35
125
34
134
25
135
24
∃ simple formula for f λ (Frame-Robinson-Thrall
hook-length formula)
Increasing and Decreasing Subsequences – p. 10
f
λ
f λ = # of SYT of shape λ
E.g., f (3,2) = 5:
123
45
124
35
125
34
134
25
135
24
∃ simple formula for f λ (Frame-Robinson-Thrall
hook-length formula)
Note. f λ = dim(irrep. of Sn ), where Sn is the
symmetric group of all permutations of
1, 2 . . . , n.
Increasing and Decreasing Subsequences – p. 10
RSK algorithm
RSK algorithm: a bijection
rsk
w → (P, Q),
where w ∈ Sn and P, Q are SYT of the same
shape λ ⊢ n.
Write λ = sh(w), the shape of w.
Increasing and Decreasing Subsequences – p. 11
RSK algorithm
RSK algorithm: a bijection
rsk
w → (P, Q),
where w ∈ Sn and P, Q are SYT of the same
shape λ ⊢ n.
Write λ = sh(w), the shape of w.
R = Gilbert de Beauregard Robinson
S = Craige Schensted (= Ea Ea)
K = Donald Ervin Knuth
Increasing and Decreasing Subsequences – p. 11
RSK algorithm
RSK algorithm: a bijection
rsk
w → (P, Q),
where w ∈ Sn and P, Q are SYT of the same
shape λ ⊢ n.
Write λ = sh(w), the shape of w.
R = Gilbert de Beauregard Robinson
S = Craige Schensted (= Ea Ea)
K = Donald Ervin Knuth
ea.ea.home.mindspring.com
Increasing and Decreasing Subsequences – p. 11
Example of RSK
w = 4132:
4
1
3
2
4
3
2
1
4
1
4
13
4
12
3
4
12
3
12
1
φ
Increasing and Decreasing Subsequences – p. 12
Example of RSK
w = 4132:
4
1
3
2
4
3
2
1
4
1
4
13
4
12
3
4
12
3
12
1
φ
(P, Q) =
12
3 ,
4
13
2
4
!
Increasing and Decreasing Subsequences – p. 12
Schensted’s theorem
rsk
Theorem. Let w → (P, Q), where
sh(P ) = sh(Q) = λ. Then
is(w) = longest row length = λ1
ds(w) = longest column length = λ′1 .
Increasing and Decreasing Subsequences – p. 13
Schensted’s theorem
rsk
Theorem. Let w → (P, Q), where
sh(P ) = sh(Q) = λ. Then
is(w) = longest row length = λ1
ds(w) = longest column length = λ′1 .
!
13
12
rsk
Example. 4132 → 3 ,
2
4
4
is(w) = 2, ds(w) = 3.
Increasing and Decreasing Subsequences – p. 13
Erdős-Szekeres theorem
Corollary (Erdős-Szekeres, Seidenberg). Let
w ∈ Spq+1 . Then either is(w) > p or ds(w) > q.
Increasing and Decreasing Subsequences – p. 14
Erdős-Szekeres theorem
Corollary (Erdős-Szekeres, Seidenberg). Let
w ∈ Spq+1 . Then either is(w) > p or ds(w) > q.
Proof. Let λ = sh(w). If is(w)
≤ p and ds(w) ≤ q
P
λi ≤ pq. then λ1 ≤ p and λ′1 ≤ q, so
Increasing and Decreasing Subsequences – p. 14
An extremal case
Corollary. Say p ≤ q. Then
#{w ∈ Spq : is(w) = p, ds(w) = q}
q 2
= f (p )
Increasing and Decreasing Subsequences – p. 15
An extremal case
Corollary. Say p ≤ q. Then
#{w ∈ Spq : is(w) = p, ds(w) = q}
q 2
= f (p )
By hook-length formula, this is
(pq)!
11 22 · · · pp (p + 1)p · · · q p (q + 1)p−1 · · · (p + q − 1)1
2
.
Increasing and Decreasing Subsequences – p. 15
Romik’s theorem
Romik: let
w ∈ Sn2 , is(w) = ds(w) = n.
Let Pw be the permutation matrix of w with
corners (±1, ±1). Then (informally) as n → ∞
almost surely the 1’s in Pw will become dense in
the region bounded by the curve
2
2 2
2
2
(x − y ) + 2(x + y ) = 3,
and will remain isolated outside this region.
Increasing and Decreasing Subsequences – p. 16
An example
w = 9, 11, 6, 14, 2, 10, 1, 5, 13, 3, 16, 8, 15, 4, 12, 17
Increasing and Decreasing Subsequences – p. 17
2
2 2
2
2
(x − y ) + 2(x + y ) = 3
1
y
–1
0.5
–0.5
0.5
1
x
–0.5
–1
Increasing and Decreasing Subsequences – p. 18
Area enclosed by curve
α = 8
Z
1
1
p
(1 − t2 )(1 − (t/3)2 )
Z 1r
1 − (t/3)2
dt
−6
2
1−t
0
0
dt
= 4(0.94545962 · · · )
Increasing and Decreasing Subsequences – p. 19
Expectation of is(w)
E(n) = expectation of is(w), w ∈ Sn
1 X
=
is(w)
n!
w∈Sn
1 X
λ 2
=
λ1 f
n!
λ⊢n
Increasing and Decreasing Subsequences – p. 20
Expectation of is(w)
E(n) = expectation of is(w), w ∈ Sn
1 X
=
is(w)
n!
w∈Sn
1 X
λ 2
=
λ1 f
n!
λ⊢n
Ulam: what is distribution of is(w)? rate of growth
of E(n)?
Increasing and Decreasing Subsequences – p. 20
Work of Hammersley
Hammersley (1972):
∃ c = lim n−1/2 E(n),
n→∞
and
π
≤ c ≤ e.
2
Increasing and Decreasing Subsequences – p. 21
Work of Hammersley
Hammersley (1972):
∃ c = lim n−1/2 E(n),
n→∞
and
π
≤ c ≤ e.
2
Conjectured c = 2.
Increasing and Decreasing Subsequences – p. 21
c=2
Logan-Shepp, Vershik-Kerov (1977): c = 2
Increasing and Decreasing Subsequences – p. 22
c=2
Logan-Shepp, Vershik-Kerov (1977): c = 2
Idea of proof.
1 X
λ 2
E(n) =
λ1 f
n!
λ⊢n
1
λ 2
.
max λ1 f
≈
n! λ⊢n
Find “limiting shape” of λ ⊢ n maximizing λ as
n → ∞ using hook-length formula.
Increasing and Decreasing Subsequences – p. 22
The limiting curve
2
1.5
1
0.5
0
0.5
1
1.5
2
Increasing and Decreasing Subsequences – p. 23
Equation of limiting curve
x = y + 2 cos θ
2
y = (sin θ − θ cos θ)
π
0≤θ≤π
Increasing and Decreasing Subsequences – p. 24
is(w) ≤ 2
uk (n) := #{w ∈ Sn : isn (w) ≤ k}.
Increasing and Decreasing Subsequences – p. 25
is(w) ≤ 2
uk (n) := #{w ∈ Sn : isn (w) ≤ k}.
J. M. Hammersley (1972):
1
2n
u2 (n) = Cn =
,
n+1 n
a Catalan number.
Increasing and Decreasing Subsequences – p. 25
is(w) ≤ 2
uk (n) := #{w ∈ Sn : isn (w) ≤ k}.
J. M. Hammersley (1972):
1
2n
u2 (n) = Cn =
,
n+1 n
a Catalan number.
For ≥160 combinatorial interpretations of Cn , see
www-math.mit.edu/∼rstan/ec
Increasing and Decreasing Subsequences – p. 25
Gessel’s theorem
I. Gessel (1990):
k
x2n
uk (n) 2 = det I|i−j| (2x) i,j=1 ,
n!
n≥0
X
where
Im (2x) =
X
j≥0
xm+2j
,
j!(m + j)!
a hyperbolic Bessel function of the first kind of
order m.
Increasing and Decreasing Subsequences – p. 26
The case k = 2
x2n
u2 (n) 2
Example.
n!
n≥0
X
= I0 (2x)2 − I1 (2x)2
x2n
=
Cn 2 .
n!
n≥0
X
Increasing and Decreasing Subsequences – p. 27
Painlevé II equation
Baik-Deift-Johansson:
Define u(x) by
d2
3
u(x)
=
2u(x)
+ xu(x) (∗),
2
dx
with certain initial conditions.
Increasing and Decreasing Subsequences – p. 28
Painlevé II equation
Baik-Deift-Johansson:
Define u(x) by
d2
3
u(x)
=
2u(x)
+ xu(x) (∗),
2
dx
with certain initial conditions.
(∗) is the Painlevé II equation (roughly, the
branch points and essential singularities are
independent of the initial conditions).
Increasing and Decreasing Subsequences – p. 28
Paul Painlevé
1863: born in Paris.
Increasing and Decreasing Subsequences – p. 29
Paul Painlevé
1863: born in Paris.
1890: Grand Prix des Sciences Mathématiques
Increasing and Decreasing Subsequences – p. 29
Paul Painlevé
1863: born in Paris.
1890: Grand Prix des Sciences Mathématiques
1908: first passenger of Wilbur Wright; set flight
duration record of one hour, 10 minutes.
Increasing and Decreasing Subsequences – p. 29
Paul Painlevé
1863: born in Paris.
1890: Grand Prix des Sciences Mathématiques
1908: first passenger of Wilbur Wright; set flight
duration record of one hour, 10 minutes.
1917, 1925: Prime Minister of France.
Increasing and Decreasing Subsequences – p. 29
Paul Painlevé
1863: born in Paris.
1890: Grand Prix des Sciences Mathématiques
1908: first passenger of Wilbur Wright; set flight
duration record of one hour, 10 minutes.
1917, 1925: Prime Minister of France.
1933: died in Paris.
Increasing and Decreasing Subsequences – p. 29
The Tracy-Widom distribution
Z
F (t) = exp −
∞
t
(x − t)u(x)2 dx
where u(x) is the Painlevé II function.
Increasing and Decreasing Subsequences – p. 30
The Baik-Deift-Johansson theorem
Let χ be a random variable with distribution F ,
and let χn be the random variable on Sn :
√
isn (w) − 2 n
χn (w) =
.
1/6
n
Increasing and Decreasing Subsequences – p. 31
The Baik-Deift-Johansson theorem
Let χ be a random variable with distribution F ,
and let χn be the random variable on Sn :
√
isn (w) − 2 n
χn (w) =
.
1/6
n
Theorem. As n → ∞,
χn → χ
in distribution,
i.e.,
lim Prob(χn ≤ t) = F (t).
n→∞
Increasing and Decreasing Subsequences – p. 31
Expectation redux
√
Recall E(n) ∼ 2 n.
Increasing and Decreasing Subsequences – p. 32
Expectation redux
√
Recall E(n) ∼ 2 n.
Corollary to BDJ theorem.
Z
√
E(n) = 2 n +
t dF (t) n1/6 + o(n1/6 )
√
= 2 n − (1.7711 · · · )n1/6 + o(n1/6 )
Increasing and Decreasing Subsequences – p. 32
Proof of BDJ theorem
Gessel’s theorem reduces the problem to “just”
analysis, viz., the Riemann-Hilbert problem in
the theory of integrable systems, and the
method of steepest descent to analyze the
asymptotic behavior of integrable systems.
Increasing and Decreasing Subsequences – p. 33
Origin of Tracy-Widom distribution
Where did the Tracy-Widom distribution F (t)
come from?
Z ∞
F (t) = exp −
(x − t)u(x)2 dx
t
d2
3
u(x)
=
2u(x)
+ xu(x)
2
dx
Increasing and Decreasing Subsequences – p. 34
Gaussian Unitary Ensemble (GUE)
Analogue of normal distribution for n × n
hermitian matrices M = (Mij ):
Increasing and Decreasing Subsequences – p. 35
Gaussian Unitary Ensemble (GUE)
Analogue of normal distribution for n × n
hermitian matrices M = (Mij ):
−1 −tr(M 2 )
Zn e
dM,
dM =
Y
i
dMii ·
Y
d(ℜ Mij )d(ℑ Mij ),
i<j
where Zn is a normalization constant.
Increasing and Decreasing Subsequences – p. 35
Tracy-Widom theorem
Tracy-Widom (1994): let α1 denote the largest
eigenvalue of M . Then
lim
n→∞
√ √ 1/6
2n ≤ t
Prob α1 − 2n
= F (t).
Increasing and Decreasing Subsequences – p. 36
Random topologies
Is the connection between is(w) and GUE a
coincidence?
Increasing and Decreasing Subsequences – p. 37
Random topologies
Is the connection between is(w) and GUE a
coincidence?
Okounkov provides a connection, via the theory
of random topologies on surfaces. Very briefly,
a surface can be described in two ways:
Gluing polygons along their edges, connected
to random matrices via quantum gravity.
Ramified covering of a sphere, which can be
formulated in terms of permutations.
Increasing and Decreasing Subsequences – p. 37
Two variations
1. Matchings
2. Alternating subsequences
Increasing and Decreasing Subsequences – p. 38
Collaborators
Joint with:
ª¡
Eva Deng ¢«¦
Rosena Du £§¨
Catherine Yan ©¥¤
Bill Chen
Increasing and Decreasing Subsequences – p. 39
Complete matchings
(complete) matching:
Increasing and Decreasing Subsequences – p. 40
Complete matchings
(complete) matching:
total number of matchings on
[2n] := {1, 2, . . . , 2n} is
(2n − 1)!! := 1 · 3 · 5 · · · (2n − 1).
Increasing and Decreasing Subsequences – p. 40
Crossings and nestings
3−crossing
3−nesting
Increasing and Decreasing Subsequences – p. 41
Crossing and nesting number
M = matching
cr(M ) = max{k : ∃ k-crossing}
ne(M ) = max{k : ∃ k-nesting}
Increasing and Decreasing Subsequences – p. 42
Crossing and nesting number
M = matching
cr(M ) = max{k : ∃ k-crossing}
ne(M ) = max{k : ∃ k-nesting}
Theorem. The number of matchings on [2n] with
no crossings (or with no nestings) is
1
2n
Cn :=
.
n+1 n
Increasing and Decreasing Subsequences – p. 42
Main result on matchings
Theorem. Let fn (i, j) = # matchings M on [2n]
with cr(M ) = i and ne(M ) = j. Then
fn (i, j) = fn (j, i).
Increasing and Decreasing Subsequences – p. 43
Main result on matchings
Theorem. Let fn (i, j) = # matchings M on [2n]
with cr(M ) = i and ne(M ) = j. Then
fn (i, j) = fn (j, i).
Corollary. # matchings M on [2n] with cr(M ) = k
equals # matchings M on [2n] with ne(M ) = k.
Increasing and Decreasing Subsequences – p. 43
Oscillating tableaux
φ
shape (3, 1), length 8
Increasing and Decreasing Subsequences – p. 44
Oscillating tableaux
φ
shape (3, 1), length 8
M
φ
Φ( M )
φ
4
2
4
3
1
3
2
1
4
2
4
2
23
13
2
1
2
1
φ
φ
Increasing and Decreasing Subsequences – p. 44
Proof sketch
Φ is a bijection from matchings on 1, 2, . . . , 2n to
oscillating tableaux of length 2n, shape ∅.
Corollary. Number of oscillating tableaux of
length 2n, shape ∅, is (2n − 1)!!.
Increasing and Decreasing Subsequences – p. 45
Proof sketch
Φ is a bijection from matchings on 1, 2, . . . , 2n to
oscillating tableaux of length 2n, shape ∅.
Corollary. Number of oscillating tableaux of
length 2n, shape ∅, is (2n − 1)!!.
(related to Brauer algebra of dimension
(2n − 1)!!).
Increasing and Decreasing Subsequences – p. 45
Schensted for matchings
Schensted’s theorem for matchings. Let
Φ(M ) = (∅ = λ0 , λ1 , . . . , λ2n = ∅).
Then
cr(M ) = max{(λi )′1 : 0 ≤ i ≤ n}
ne(M ) = max{λi1 : 0 ≤ i ≤ n}.
Increasing and Decreasing Subsequences – p. 46
Schensted for matchings
Schensted’s theorem for matchings. Let
Φ(M ) = (∅ = λ0 , λ1 , . . . , λ2n = ∅).
Then
cr(M ) = max{(λi )′1 : 0 ≤ i ≤ n}
ne(M ) = max{λi1 : 0 ≤ i ≤ n}.
Proof. Reduce to ordinary RSK.
Increasing and Decreasing Subsequences – p. 46
M
′
Now let cr(M ) = i, ne(M ) = j, and
Φ(M ) = (∅ = λ0 , λ1 , . . . , λ2n = ∅).
Define M ′ by
Φ(M ′ ) = (∅ = (λ0 )′ , (λ1 )′ , . . . , (λ2n )′ = ∅).
By Schensted’s theorem for matchings,
cr(M ′ ) = j, ne(M ′ ) = i.
Increasing and Decreasing Subsequences – p. 47
Conclusion of proof
Thus M 7→ M ′ is an involution on matchings of
[2n] interchanging cr and ne.
⇒ Theorem. Let fn (i, j) = # matchings M on
[2n] with cr(M ) = i and ne(M ) = j. Then
fn (i, j) = fn (j, i).
Increasing and Decreasing Subsequences – p. 48
Simple description?
Open: simple description of M 7→ M ′ , the
analogue of
a1 a2 · · · an 7→ an · · · a2 a1 ,
which interchanges is and ds.
Increasing and Decreasing Subsequences – p. 49
gk (n)
gk (n) = number of matching M on [2n]
with cro(M ) ≤ k
(matching analogue of uk (n))
Increasing and Decreasing Subsequences – p. 50
Grabiner-Magyar theorem
Theorem. Define
Hk (x) =
X
n
Then
where
x2n
.
gk (n)
(2n)!
k
Hk (x) = det I|i−j| (2x) − Ii+j (2x) i,j=1
Im (2x) =
X
j≥0
xm+2j
j!(m + j)!
as before.
Increasing and Decreasing Subsequences – p. 51
Noncrossing example
Example. k = 1 (noncrossing matchings):
H1 (x) = I0 (2x) − I2 (2x)
X
x2j
=
Cj
.
(2j)!
j≥0
Increasing and Decreasing Subsequences – p. 52
Baik-Rains theorem
Baik-Rains (implicitly):
lim Prob
n→∞
√
crn (M ) − 2n
t
≤
1/6
2
(2n)
!
= F1 (t),
where
Z ∞
p
1
F1 (t) = F (t) exp
u(s)ds ,
2 t
where F (t) is the Tracy-Widom distribution and
u(t) the Painlevé II function.
Increasing and Decreasing Subsequences – p. 53
Alternating permutations
Alternating sequence of length k:
b1 > b2 < b3 > b4 < · · · bk
En : number of alternating w ∈ Sn (Euler
number)
E4 = 5: 2134, 3142, 3241, 4132, 4231
Increasing and Decreasing Subsequences – p. 54
Alternating permutations
Alternating sequence of length k:
b1 > b2 < b3 > b4 < · · · bk
En : number of alternating w ∈ Sn (Euler
number)
E4 = 5: 2134, 3142, 3241, 4132, 4231
Desirée André (1879):
xn
En = sec x + tan x
n!
n≥0
X
Increasing and Decreasing Subsequences – p. 54
Alternating subsequences?
as(w) = length of longest alternating subseq. of w
Increasing and Decreasing Subsequences – p. 55
Alternating subsequences?
as(w) = length of longest alternating subseq. of w
w = 56218347 ⇒ as(w) = 5
Increasing and Decreasing Subsequences – p. 55
The main lemma
MAIN LEMMA. ∀ w ∈ Sn ∃ alternating
subsequence of maximal length that contains n.
Increasing and Decreasing Subsequences – p. 56
The main lemma
MAIN LEMMA. ∀ w ∈ Sn ∃ alternating
subsequence of maximal length that contains n.
ak (n) = #{w ∈ Sn : as(w) = k}
bk (n) = a1 (n) + a2 (n) + · · · + ak (n)
= #{w ∈ Sn : as(w) ≤ k}.
Increasing and Decreasing Subsequences – p. 56
Recurrence for ak (n)
⇒ ak (n) =
X
2r+s=k−1
n X
n−1
j=1
j−1
(a2r (j − 1) + a2r+1 (j − 1)) as (n − j)
Increasing and Decreasing Subsequences – p. 57
The main generating function
Define B(x, t) =
X
k,n≥0
n
x
bk (n)tk
n!
Theorem.
2/ρ
1
B(x, t) =
1−ρ ρx − ρ ,
1− t e
√
where ρ= 1 − t2 .
Increasing and Decreasing Subsequences – p. 58
Formulas for bk (n)
Corollary.
⇒ b1 (n) = 1
b2 (n) = n
b3 (n) = 14 (3n − 2n + 3)
b4 (n) =
..
.
1 n
8 (4
− (2n − 4)2n )
Increasing and Decreasing Subsequences – p. 59
Formulas for bk (n)
Corollary.
⇒ b1 (n) = 1
b2 (n) = n
b3 (n) = 14 (3n − 2n + 3)
b4 (n) =
..
.
1 n
8 (4
− (2n − 4)2n )
no such formulas for longest increasing
subsequences
Increasing and Decreasing Subsequences – p. 59
Mean (expectation) of as(w)
1 X
A(n) =
as(w),
n!
w∈Sn
the expectation of as(n) for w ∈ Sn
Increasing and Decreasing Subsequences – p. 60
Mean (expectation) of as(w)
1 X
A(n) =
as(w),
n!
w∈Sn
the expectation of as(n) for w ∈ Sn
√
Recall E(n) ∼ 2 n.
Increasing and Decreasing Subsequences – p. 60
Mean (expectation) of as(w)
1 X
A(n) =
as(w),
n!
w∈Sn
the expectation of as(n) for w ∈ Sn
√
Recall E(n) ∼ 2 n.
Corollary.
4n + 1
A(n) =
, n≥2
6
Increasing and Decreasing Subsequences – p. 60
Variance of as(w)
2
X
4n + 1
1
as(w) −
V (n) =
, n≥2
n!
6
w∈Sn
the variance of as(n) for w ∈ Sn
Increasing and Decreasing Subsequences – p. 61
Variance of as(w)
2
X
4n + 1
1
as(w) −
V (n) =
, n≥2
n!
6
w∈Sn
the variance of as(n) for w ∈ Sn
Corollary.
8
13
V (n) = n −
, n≥4
45
180
Increasing and Decreasing Subsequences – p. 61
Variance of as(w)
2
X
4n + 1
1
as(w) −
V (n) =
, n≥2
n!
6
w∈Sn
the variance of as(n) for w ∈ Sn
Corollary.
8
13
V (n) = n −
, n≥4
45
180
similar results for higher moments
Increasing and Decreasing Subsequences – p. 61
A new distribution?
P (t) = lim Probw∈Sn
n→∞
asn (w) − 2n/3
√
≤t
n
Increasing and Decreasing Subsequences – p. 62
A new distribution?
P (t) = lim Probw∈Sn
n→∞
asn (w) − 2n/3
√
≤t
n
Stanley distribution?
Increasing and Decreasing Subsequences – p. 62
Limiting distribution
Theorem (Pemantle, Widom, (Wilf)).
as(w) − 2n/3
√
lim Probw∈Sn
≤t
n→∞
n
1
=√
π
(Gaussian distribution)
Z
√
t 45/4
e
−s2
ds
−∞
Increasing and Decreasing Subsequences – p. 63
Limiting distribution
Theorem (Pemantle, Widom, (Wilf)).
as(w) − 2n/3
√
lim Probw∈Sn
≤t
n→∞
n
1
=√
π
(Gaussian distribution)
Z
√
t 45/4
e
−s2
ds
−∞
Increasing and Decreasing Subsequences – p. 63
Increasing and Decreasing Subsequences – p. 64