iii proceedings of iii congress struga, macedonia, 29.ix-2 ...math.bilten.smm.com.mk/source/iii...
TRANSCRIPT
225
Зборник на трудови од III конгрес на математичарите на Македонија Струга, Македонија, 29.IX-2.X-2005
Стр. 225-238
Proceedings of III congress of mathematicians of Macedonia
Struga, Macedonia, 29.IX-2.X-2005 Pages 225-238
QUASICOMPOSITIONS OF NATURAL NUMBER
Adi Dani1
In this paper number 0 is included in set of natural numbers. By
{ }0,1,..., 1mI m= − is denoted the set of all natural numbers lesser than given natural number m. Definition 1. Each m sequence ( )0 1 1 1
, ,..., m kc c c c − +
= with terms in set
{ }1 0,1,...,kI k+ = with condition 0 1 1... mc c c k−+ + + = is called quasicomposition of natural number k in m parts. Theorem 1. Each m quasicomposition of number k defines a k nondecreasing (or nonincreasing) sequence with terms in mI end vice versa. Proof. Show the biective correspondence below
( )0 11
0 1 1 1, ,..., 0,0,...0,1,1,...,1,..., 1, 1,..., 1
m
m kc c c
m
c c c m m m−
− +
⎛ ⎞⎜ ⎟↔ − − −⎜ ⎟⎝ ⎠
Theorem 2. Each nondecreasing k sequence with terms in Im defines a k subset of 1m kI + − and vice versa.
Proof. Let be ( )0 1 1, ,..., k mv v v v −= a nondecreasing k sequence with terms in mI
following definition we have ( )
( )1
, 0 1 ( 1)
(0 2)k i m i i
i i m k
i I v I v m i v i m i
v i m k v i I + −
∀ ∈ ∈ ⇒ ≤ ≤ − ⇒ ≤ + ≤ + − ⇒
⇒ ≤ + ≤ + − ⇒ + ∈
that mean sequence v define (determine) set ( ) { } 1:i k m kI v v i i I I + −= + ∈ ⊆
clearly ( )I v k=
because( ), , , 0,k i j i j i j i ji j I i j v v j i v v j i v i v j v i v j∀ ∈ < ≤ ⇒ − > < + − ⇒ + < + ⇒ + ≠ +
Conversely: Let be { }0 1 1 1, ,..., k m kB b b b I− + −= ⊆ a k set then we can suppose that
( )0 1 10 ... 2 1 1kb b b m k m k−≤ < < < ≤ + − = + − −
then 1 0 1k i i i mi I i b m i b i m b i I∀ ∈ ⇒ ≤ ≤ + − ⇒ ≤ − ≤ − ⇒ − ∈
A. Dani
226
1 1 1 1i i i i i ib b b i b i b i b i+ + +< ⇒ − < − ⇒ − ≤ − − that mean set B defines a nondecreasing k sequence
( ) ( )( )0 1 10, 1,..., 1k mv B b b b k−= − − − −
who’s terms are in set mI .
Denote by ( )mq k the number of all quasicompositions of number k in m-parts.
According to theorem 1 and theorem 2 that number is equal at the number 1m k
k+ −⎛ ⎞
⎜ ⎟⎝ ⎠
of all k subsets of set 1m kI + − . We can resume
[1] ( )0 1 1 0 1 1 0 1 1... 0 ... 1 0 ... 2
11 1 1
m k k
mc c c k v v v m b b b m k
m kq k
k− − −+ + = ≤ ≤ ≤ ≤ ≤ − ≤ < < < ≤ + −
+ −⎛ ⎞= = = = ⎜ ⎟
⎝ ⎠∑ ∑ ∑
for example.
( )3
3 4 1 6 6 6 54 154 4 2 2 1
q+ −⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⋅= = = = =⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⋅⎝ ⎠ ⎝ ⎠ ⎝ ⎠
that mean there exists exactly 15 • Quasicompositions of number 4 in 3 parts • No decreasing 4 sequences with terms in 3I • 4 subsets of set 6I
In extension, we present all those configurations by colons ( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )
5 3 65 3 65 3 65 3 65 3 6
0,0, 4 2,2,2,2 2,3,4,50,1,3 1,2,2,2 1,3,4,50,2,2 1,1,2,2 1,2,4,50,3,1 1,1,1,2 1,2,3,50,4,0 1,1,1,1 1,2,3,4
↔ ↔↔ ↔↔ ↔↔ ↔↔ ↔
( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )
( ) ( ) ( )
5 3 6
5 3 6
5 3 6
5 3 6
5 3 6
1,0,3 0,2,2,2 0,3,4,5
1,1,2 0,1,2,2 0, 2,4,5
1,2,1 0,1,1,2 0,2,3,5
1,3,0 0,1,1,1 0, 2,3,4
2,0, 2 0,0,2,2 0,1,4,5
↔ ↔
↔ ↔
↔ ↔
↔ ↔
↔ ↔
( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )
5 3 6
5 3 6
5 3 6
5 3 6
5 3 6
2,1,1 0,0,1,2 0,1,3,5
2,2,0 0,0,1,1 0,1,3,4
3,0,1 0,0,0,2 0,1,2,5
3,1,0 0,0,0,1 0,1,2,4
4,0,0 0,0,0,0 0,1,2,3
↔ ↔
↔ ↔
↔ ↔
↔ ↔
↔ ↔
Quasicompositions of natural number
227
Each quasicomposition ( )0 1 1 1, ,..., m k
c c c c − += of number k in m parts explicitly we can
present as array of boxes with common borders with objects placed in boxes in i-th box we put ic objects (dots) mi I∀ ∈
0 1 1
... ... ... ...mc c c −
ii i ii i ii i
That scheme of boxes we call putting of k similar objects in m different boxes. For example
( )41, 2,0,3 ↔ i ii iii
Theorem 3. The number ( )mq k satisfy the following relations
( )( )( ) ( ) ( )
0
1
1. 0
2. 0 1, 0
3. 1 , 0, 0m
m m m
q k
q m
q k q k q k m k−
=
= >
= + − > >
Proof. The first two equations are evident, third equation is true because the set of all quasicompositions of k in m parts can be parted in two disjoint set. One set contain all quasicompositions whose first term (part) is 0 and the other set contain the rest of quasicompositions. Each quasicomposition of first set defines a quasicomposition of k in 1m − parts; in other hand, each quasicomposition of second set defines a quasicomposition of 1k − in m parts. The numbers ( )mq k can be placed in an infinite table with rows m and columns k
• All elements of first row 0m = in accordance with 1. are 0 • First element of first column k = 0 is 0 others elements in accordance with
2. are 1 • Every element of table that is not in first column or in first row in
accordance with recurrence 3. is a sum of his neighbor on left with his neighbor up to him. \ 0 1 2 3 4 5 6 7 8 9 .0 0 0 0 0 0 0 0 0 0 0 .1 1 1 1 1 1 1 1 1 1 1 .2 1 2 3 4 5 6 7 8 9 10 .3 1 3 6 10 15 21 28 36 45 55 .4 1 4 10 20 35 56 84 120 165 220 .5 1 5 15 35 70 126 210 330 495 715 .6 1 6 21 56 126 252 462 792 1287 2002 .7 1 7 28 84 210 462 924 1716 3003 5005 .8 1 8 36 120 330 792 1716 3432 643
m k
5 11440 .9 1 9 45 165 495 1287 3003 6435 12870 24310 .. . . . . . . . . . . .
Table 1
A. Dani
228
Definition. The formal power series ( ) ( )0
km m
kq x q k x
∞
=
=∑
is called generating function of number of quasicompositions of number k in m parts. From
[2] ( )0 1 1...
11
m
mc c c k
m kq k
k−+ + + =
+ −⎛ ⎞= = ⎜ ⎟
⎝ ⎠∑
we have
[3] ( ) ( )0 11
0 1 1 0 1 10 0 ... 0 0 0
... 1m
m m
mc cck km
k k c c c k c c c
q k x x x x x x−
− −
∞ ∞ ∞ ∞ ∞−
= = + + + = = = =
= = = −∑ ∑ ∑ ∑ ∑ ∑
Conclusion: if 1x < then
[4] ( )0
11 m k
k
m kx x
k
∞−
=
+ −⎛ ⎞− = ⋅⎜ ⎟
⎝ ⎠∑
Definition. Let be ( )0 1 1 1, ,..., m k
c c c c − += a quasicomposition of k in m parts then
vector ( ) ( )0 1 1
, ,..., k mg c t t t
+= where 1,i kt i I +∀ ∈ express the number of coordinates of c
those are equal at i, is called trace of quasicomposition c. Obviously the trace satisfy the conditions
1 2
0 1
1 2 ......
k
k
t t kt kt t t m
+ + + =+ + + =
Denote by ( )0 1 1, ,..., k m
P t t t+
the set of all quasicompositions c of number k in m parts
with same trace ( ) ( )0 1 1, ,..., k m
g c t t t+
= and let be ( )0 1 1 1, ,..., k m
c P t t t − +∈ then
,i kt i I∀ ∈ is the number of coordinates of c who’s are equal at i, if we rearrange them in !it ways then all m coordinates of c can be arranged in 0 1! !... !kt t t ways. All of !m
possible arrangements of m coordinates of quasicompositions of ( )0 1 1, ,..., k m
P t t t+
we
can get, if that process repeat for each element of set ( )0 1 1, ,..., k m
P t t t+
that give equation
( )0 1 0 11, ,..., ! !... ! !k km
P t t t t t t m+
= or ( )0 1 10 1
!, ,...,! !... !k m
k
mP t t tt t t+
=
finally if this process apply over all traces of quasicompositions of number k in m parts we get the formula
[5] ( )1 2
0 1
1 2 ... 0 1 ...
!! !... !
kk
mt t kt k kt t t m
mq kt t t+ + + =
+ + + =
= ∑
Quasicompositions of natural number
229
Quasicompositions of natural number with upper homogenous restrictions Definition. Each quasicomposition ( )0 1 1 1
, ,..., m kr r r − +
of k in m parts whose terms
fulfill the condition ,i mr s i I≤ ∀ ∈ is called quasicomposition of number k in m parts no greater than s or simply m quasicomposition of k of order s. Each m quasicomposition of k of order s define a putting of k similar objects in m different boxes and in each box we can put at most s objects Denote by
[[6]] 0 1 1...
,
1m
i m
r r r ksr s i I
mk
−+ + + =≤ ∀ ∈
⎛ ⎞=⎜ ⎟
⎝ ⎠∑
number of all m quasicomposition of k of order s, then we can prove the following equations
[1s] 0
10 s
⎛ ⎞=⎜ ⎟
⎝ ⎠
[2s] 0,s
mk ms
k⎛ ⎞
= >⎜ ⎟⎝ ⎠
[3s] min ,
0
1k s
is s
m mk k i=
−⎛ ⎞ ⎛ ⎞=⎜ ⎟ ⎜ ⎟−⎝ ⎠ ⎝ ⎠∑
[4s] s s
m mk ms k
⎛ ⎞ ⎛ ⎞=⎜ ⎟ ⎜ ⎟−⎝ ⎠ ⎝ ⎠
Proof.
0 1 1... 0 0 ,
1 10
mi m
c c csc s i I
m
−+ + + =≤ ≤ ∀ ∈
⎛ ⎞= =⎜ ⎟
⎝ ⎠∑
0 1 1... 0 ,
1 0,m
i m
c c c ksc s i I
mk ms
k−+ + + =
≤ ≤ ∀ ∈
⎛ ⎞= = >⎜ ⎟
⎝ ⎠∑
0 1 1 1 0 1 2 1 0 1 2 0 1 21 1
... 0 ... 0 ... 1 ... 0 , 0 , 0 , 0 ,
1 1 1 1
1
m m m m m mi m i m i m i m
s k s
c c c k c c c c k c i c c c k i i k c c c k isc s i I c s i I c s i I c s i I
mk
mk i
− − − − − −− −
+ + + = = + + + = − = + + + = − = + + + + = −≤ ≤ ∀ ∈ ≤ ≤ ∀ ∈ ≤ ≤ ∀ ∈ ≤ ≤ ∀ ∈
⎛ ⎞= = = + =⎜ ⎟
⎝ ⎠
−⎛=
−⎝
∑ ∑ ∑ ∑ ∑ ∑ ∑
0,0
k
i s
k s=
⎞< <⎜ ⎟
⎠∑
A. Dani
230
0 1 1 1 0 1 2 1 0 1 21
... 0 ... 0 ... 0 , 0 , 0 ,
0
1 1 1
1 ,
m m m m mi m i m i m
s s
c c c k c c c c k c i c c c k isc s i I c s i I c s i I
s
i s
mk
ms k ms
k i
− − − − −−
+ + + = = + + + = − = + + + = −≤ ≤ ∀ ∈ ≤ ≤ ∀ ∈ ≤ ≤ ∀ ∈
=
⎛ ⎞= = = =⎜ ⎟
⎝ ⎠
−⎛ ⎞= < ≤⎜ ⎟−⎝ ⎠
∑ ∑ ∑ ∑ ∑
∑
0 1 1 0 1 1 0 1 1... ( ) ( ) ... ( ) ... 0 , 0 , 0 ,
1 1 1
m m mi m i m i m
c c c k s c s c s c ms k t t t ms ksc s i I s c s i I t s i I
s
mk
mms k
− − −+ + + = − + − + + − = − + + + = −≤ ≤ ∀ ∈ ≤ − ≤ ∀ ∈ ≤ ≤ ∀ ∈
⎛ ⎞= = = =⎜ ⎟
⎝ ⎠
⎛ ⎞= ⎜ ⎟−⎝ ⎠
∑ ∑ ∑
Definition. Formal potential series
( ),0 0
msk k
m sk ks s
m mq x x x
k k
∞
= =
⎛ ⎞ ⎛ ⎞= =⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠∑ ∑
is generating function of numbers of m quasicompositions of order s considering [6] we conclude
[7] ( )0 11
0 1 1 0 1 10 0 ... 0 0 0 ,
... 1 ...m
m mi m
ms ms s s s mc cck k s
k k c c c k c c csc s i I
mx x x x x x x
k−
− −= = + + + = = = =≤ ∀ ∈
⎛ ⎞= = = + + +⎜ ⎟
⎝ ⎠∑ ∑ ∑ ∑ ∑ ∑
[7a] 1
0
1 , 11
msmsk
k s
m xx xk x
+
=
⎛ ⎞⎛ ⎞ −= ≠⎜ ⎟⎜ ⎟ −⎝ ⎠ ⎝ ⎠∑
that generating function is polynomial, the numbers s
mk
⎛ ⎞⎜ ⎟⎝ ⎠
are called polynomial
coefficients and [7] is called polynomial formula. If in [7] we put 1x = that give
[8] ( )0
1ms
m
k s
ms
k=
⎛ ⎞= +⎜ ⎟
⎝ ⎠∑
for 1x = − from [7a] we have
[8a] ( ) ( )( ) ( )
1
0
1 01 1
1 1 11 1 02
msmsk s
k s
mmk m
+
=
=⎧⎛ ⎞− −⎛ ⎞ ⎪− = =⎜ ⎟ ⎨⎜ ⎟ + −⎜ ⎟− − >⎝ ⎠ ⎪⎝ ⎠ ⎩
∑
Polynomial Theorem
If in [5] we put 1 2 ... 0s s kt t t+ += = = = then we get an other formula for counting the quasicompositions of order s
Quasicompositions of natural number
231
[9] ( )1 2
0 1
1 2 ... 0 1...
!! !... !
ss
sm
t t st k sst t t m
m mq kk t t t+ + + =
+ + + =
⎛ ⎞= =⎜ ⎟
⎝ ⎠∑
that fact lead to
[10]
( ) ( ) ( )
1 2
1 2 0 10 1
0 1
0 1
2 ...
0 0 2 ... ...0 1 0 1...
... 0 1
! !! ! ... ! ! ! ... !
! 1 ...! ! ... !
s
s ss
s
s
ms msr r srk k
k k r r sr k r r r ms ssr r r m
rr r s
r r r m s
m m mx x xk r r r r r r
m x xr r r
+ + +
= = + + + = + + + =+ + + =
+ + + =
⎛ ⎞= = =⎜ ⎟
⎝ ⎠
=
∑ ∑ ∑ ∑
∑
from [7] and [10] we get
[11] ( ) ( ) ( ) ( )0 1
0 1 ... 0 1
!1 ... 1 ...! ! ... !
s
s
m rr rs s
r r r m s
mx x x xr r r+ + + =
+ + + = ∑
if in [11] we put 10
,i is
ax i Ia += ∈ then we get the list of equations
0 1
0 1
01 1
...0 0 0 1 0 0 0
!1 ... ...! !... !
s
s
m r r r
s s
r r r m s
a a aa ama a r r r a a a+ + + =
⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞+ + + =⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠∑
0 1
0 10 1
0 1 0 1...
...0 0 1 0
... ...!! !... !
s
ss
m r rrs s
r r rr r r m s
a a a a a ama r r r a + + +
+ + + =
⎛ ⎞+ + + =⎜ ⎟⎝ ⎠
∑
( ) 0 1
0 1
0 1 0 1
...0 0 1 0
... ...!! !... !
s
s
m r rrs s
m mr r r m s
a a a a a ama r r r a+ + + =
+ + += ∑
[12] ( ) 0 1
0 1
0 1 0 1... 0 1
!... ...! !... !
s
s
m r rrs s
r r r m s
ma a a a a ar r r+ + + =
+ + + = ∑
the last equation is called polynomial theorem If in [9] we put 1s = then
( ) ( )1 10 1
1 0 1 1 11
! ! !! ! ! ! ! !t k t k
t t m
m mm m mk kt t m t t m k k= =
+ =
⎛ ⎞ ⎛ ⎞= = = =⎜ ⎟ ⎜ ⎟− −⎝ ⎠ ⎝ ⎠∑ ∑
If in formula, [7] put 1s = that give Newton binomial formula
[13] ( )0
1m
mk
k
mx x
k=
⎛ ⎞= +⎜ ⎟
⎝ ⎠∑
Now in [13] we suppose that axb
= then we get binomial theorem
[13a] ( )0
mm k m k
k
ma b a b
k−
=
⎛ ⎞+ = ⎜ ⎟
⎝ ⎠∑
A. Dani
232
Generalized Fibonacci sequence Let be s ∈ a fixed natural number then infinite sequence
( )(0), (1),..., ( ),...s s s sF F F F m=
who fulfill the conditions ( )0 0,F m m= ∈ for 0s >
( )
( ) ( )
( ) ( )
1
0
1
0
0 1
1 ,
1 ,
s
m
s si
s
s si
F
F m F m i m s
F m F m i m s
−
=
−
=
=
= − − <
= − − ≥
∑
∑
is called Fibonacci sequence of order s. This definition implies that the Fibonacci sequences of order 0 and 1 are constant all those terms are 0 respectively 1.
( ) ( )0 10, 1,F m F m m N= = ∀ ∈
Fibonacci sequence of order 2s = we denote by 2F F= is defined by
( ) ( ) ( ) ( ) ( )0 1 1 ; 1 2 , 2F F F m F m F m m= = = − + − ≥ this is well known usual Fibonacci sequence. In table below are listed the ten first terms of Fibonacci sequences of order no greater than 4
0
1
2
3
4
0 1 2 3 4 5 6 7 8 9 .0 0 0 0 0 0 0 0 0 0 .1 1 1 1 1 1 1 1 1 1 .1 1 2 3 5 8 13 21 34 55 .1 1 2 4 7 13 24 44 81 149 .1 1 2 4 8 15 29 56 108 208 .
. . . . . . . . . . . .
FFFFF
Definition. Formal power series ( ) ( )0
ms s
m
F x F m x∞
=
=∑
is generating function of Fibonacci sequence of order s. From definitions we have
( ) ( ) ( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( )
( )( )
0
0 1 1
11 1 1
1 1 0 0 0
1 1 11 1 1
0 0 0 0
0 1
1 1 1 1 1
1 1 1
1 ...
m m ms s s s s
m m m
sm m m
s s sm m m i
s s sm i i m i i i
s s si m i m i i
ss
F x F m x F x F m x F m x
F m x F m x F m i x
F m i x x F m i x x F x x
F x x x
∞ ∞ ∞
= = =
∞ ∞ ∞ −+ + +
+ = = = =
− ∞ − ∞ −− + − + +
= = = = =
= = + = + =
= + + = + + = + − =
⎛ ⎞= + − = + − = + =⎜ ⎟⎝ ⎠
= + + +
∑ ∑ ∑
∑ ∑ ∑∑
∑∑ ∑ ∑ ∑
Quasicompositions of natural number
233
Solve this equation by ( )sF x we have
( ) 20
11 ( ... )
ms s
mF m x
x x x
∞
=
=− + + +∑
Considering [7] then we have
( ) ( ) ( )
( )
22
0 0
1
0 0 0 0 01 1
0 0 1
1 ... 1 ...1 ...
1 ...
,
mm s ss s
m m
mm s m i m i
m m i m is s
m
m i s
F m x x x x x xx x x
m mx x x x x x
i i
m i m i mx
i m m
∞ ∞
= =
∞ ∞ ∞ ∞ ∞− +
= = = = =− −
∞ ∞
= = −
= = + + < = + + + =− + + +
⎛ ⎞ ⎛ ⎞= + + + = = =⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠− + →⎛ ⎞
= ⎜ ⎟ → −⎝ ⎠
∑ ∑
∑ ∑ ∑ ∑∑
∑∑ i
equalizing the coefficients of x with the same power, we get
[14] ( )0 01 1
si is s
m i m iF m
i i
∞
= ≥− −
− −⎛ ⎞ ⎛ ⎞= =⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠∑ ∑
Identity of Vandermonde
From polynomial formula ( )
( ) ( ) ( )0 1
0 1 0 1
0 0 01 1 10 0 1 0 11
0 1 0 1 0 1
0 1
0
0 0 01 1 1
0 0 0 0 0 00 0 01 1 1
1 ... 1 ... 1 ...a a s
a a a ak s s s
k s
a s a s a sa s a s a st t t t tt
t t t t t tss s s
a ax x x x x x x
k
a a aa a ax x x x
t t tt t t
++
=
+ +
= = = = = =
+⎛ ⎞= + + + = + + + + + + =⎜ ⎟
⎝ ⎠
⎛ ⎞ ⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞ ⎛ ⎞= = =⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠ ⎝ ⎠
∑
∑ ∑ ∑ ∑ ∑∑( )0 1
0 1
0 1
0 0 1
a a sk
k t t k ss
a ax
t t
+
= + =
=
⎛ ⎞ ⎛ ⎞= ⎜ ⎟ ⎜ ⎟
⎝ ⎠⎝ ⎠∑ ∑
from above we gate the relation so called Vandermonde identity of order s
0 1 0
0 0 110 1
00 0 01
k
t t k ts ss s s
a a aaa at t k ttk + = =
+ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞⎛ ⎞⎛ ⎞= =⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟⎜ ⎟ −⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠ ⎝ ⎠∑ ∑
denote 0 1 0 1, ka a m k ms t i I t k i= = = = ∈ ⇒ = − then considering [4.s] the last equation turn into
[15] 2
0 0 0
2 ms ms ms
i i is s s s s s
m m m m m mms i ms i i i i= = =
⎛ ⎞⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞= = = ⎜ ⎟⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠∑ ∑ ∑
in the same way we can prove that
0 1 1
0 110 1 1
... 0 11
......
m
mm
t t t k ms ss s
a aaa a at ttk
−
−−
+ + + = −
+ + + ⎛ ⎞ ⎛ ⎞⎛ ⎞⎛ ⎞= ⎜ ⎟ ⎜ ⎟⎜ ⎟⎜ ⎟
⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠∑
that is called generalized Vandermonde identity.
A. Dani
234
Generalized Pascal triangle An infinite table with rows m and columns k that in meting point of row m with column
k is placed number s
mk
⎛ ⎞⎜ ⎟⎝ ⎠
is called Pascal triangle of order s
0 10 0 0
00 1
1 1 11
0 1
0 1
s s s
s s s
s s s
k
k
k
m m mm
k
⋅ ⋅
⎛ ⎞ ⎛ ⎞ ⎛ ⎞⋅ ⋅⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠ ⎝ ⎠
⎛ ⎞ ⎛ ⎞ ⎛ ⎞⋅ ⋅⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠ ⎝ ⎠⋅ ⋅ ⋅ ⋅ ⋅ ⋅
⎛ ⎞ ⎛ ⎞ ⎛ ⎞⋅ ⋅⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠ ⎝ ⎠⋅ ⋅ ⋅ ⋅ ⋅ ⋅
Pascal triangle of order s
Basic properties of this table are. • First row 0m = begin with 1 from [1s] all other elements are 0 from [2s] • Each element who isn’t in first row following [3s] is the sum of his neighbor
up to him and min ,k s neighbors who’s are placed on left in the same row. Pascal triangle can be constructed row after row recursively. Below we make mention at 4 important properties of Pascal triangle of order s
1. The sum of elements in row m is equal at ( )1 ms + (from [8])
2. Alternative sum in row m is ( )1 12
ms⎛ ⎞+ −⎜ ⎟⎜ ⎟⎝ ⎠
(from [8a])
3. sum of squares of elements in row m is equal at 2
s
mms
⎛ ⎞⎜ ⎟⎝ ⎠
(from [14])
4. sum of elements in m –th small diagonal is equal at ( )1sF m+ (from [15]) For 0s = we have Pascal triangle of order 0 this triangle not has interesting properties, all elements of first column are 1 and all others are equal at 0. In the tables below we present the Pascal triangles of order, 1s = that is usual Pascal triangle, and Pascal triangles of orders 2s = and 3s = .
Quasicompositions of natural number
235
0 3 5 6 7 8 91 2 40 11 1 12 1 2 13 3 31 1
64 1 4 4 15 5 10 10 51 16 6 15 20 15 61 17 7 35 35 71 21 21 18 8 28 56 70 56 28 81 19 9 36 84 126 126 84 36 91 1
⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅
Pascal triangle of order 1
0 3 5 6 7 8 91 2 40 11 1 1 1
32 1 2 2 13 3 6 7 6 31 1
10 16 19 16 104 1 4 4 15 5 15 30 45 51 45 30 15 516 6 50 90 126 126 90 501 21 1417 7 28 77 161 266 357 393 357 26618 8 36 266 560 784 1016 1107 10161 1129 9 45 156 438 990 1610 1960 2907 31391
⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ Pascal triangle of order 2
A. Dani
236
0 3 5 6 7 8 9 .1 2 40 .1
.1 1 1 1 13 3 .2 1 2 4 2 1
3 3 6 10 10 6 3 .1 12 12 110 20 31 40 40 31 20 .4 1 4 44
5 5 15 35 65 101 135 155 155 135 .16 6 56 120 216 336 456 546 5801 217 7 28 84 203 413 728 1128 1554 1918 .18 8 36 120 322 728 1428 2472 3823 5328 .19 9 45 165 486 1206 2598 4950 81
⋅
451 13051 .. . . . . . . . . . . .
Pascal triangle of order 3 Theorem From [7], [13], and [5] we can conclude that number of quasicompositions of k in m parts of order s have the following properties
[16] 0 1
m
is s
m m ik i k i= −
⎛ ⎞ ⎛ ⎞⎛ ⎞=⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠∑
[17] 0 1
m
is s
m m m ik i k si= −
−⎛ ⎞ ⎛ ⎞⎛ ⎞=⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠∑
[18] ( ) ( )0
1 11
1
mi
is
m m m k s ik i m=
⎛ + − + − ⎞⎛ ⎞ ⎛ ⎞= − ⎜ ⎟⎜ ⎟ ⎜ ⎟ −⎝ ⎠ ⎝ ⎠⎝ ⎠∑
Proof: From
( ) ( )( )
( ) ( )
2
0
2 1
0 0
0 0 0 01 1
1 ... 1 ...
... 1 ...
mmk s s
k s
m mi is i s
i i
m mi j k
i j k is s
mx x x x x x
k
m mx x x x x x
i i
m i m ix x
i j i k i
∞
=
−
= =
∞ ∞+
= = = =− −
⎛ ⎞= + + + = + + + + =⎜ ⎟
⎝ ⎠
⎛ ⎞ ⎛ ⎞= + + + = + + + =⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠⎛ ⎞ ⎛ ⎞ ⎛ ⎞⎛ ⎞
= =⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠
∑
∑ ∑
∑ ∑ ∑∑
we get [16], similarly from
Quasicompositions of natural number
237
( ) ( )( )
( )
1
0
1
0 0 0 1
0 0 0 01 1
1 ... 1 ...
1 ...
mmk s s s
k s
m mm is si si j
i i j s
m msi j k
i j k is s
mx x x x x x
k
m m m ix x x x x
i i j
m m i m m ix x
i j i k si
∞−
=
∞−−
= = = −
∞ ∞+
= = = =− −
⎛ ⎞= + + + = + + + + =⎜ ⎟
⎝ ⎠−⎛ ⎞ ⎛ ⎞ ⎛ ⎞
= + + + = =⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠ ⎝ ⎠
− −⎛ ⎞ ⎛ ⎞ ⎛ ⎞⎛ ⎞= =⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠
∑
∑ ∑ ∑
∑ ∑ ∑∑
we get [17] and finally from
( ) ( )
( ) ( ) ( )
( ) ( )
( ) ( )( ) ( )
11
0
11
0 0 0 0
0 0
1 1 11
1 11
1 1
1 11 1
1
ms m mk s
k s
m mi i j s is j
i j i j
mi ik
i k
m xx x xk x
m m j m m jx x x
i j i j
j s i k j k s i
m k s im mx
k s ii i
+∞−+
=
∞ ∞+ ++
= = = =
∞
= =
⎛ ⎞⎛ ⎞ −= = − − =⎜ ⎟⎜ ⎟ −⎝ ⎠ ⎝ ⎠+ − + −⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
− = − =⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠+ + = ⇒ = − +
+ − + −⎛ ⎞⎛ ⎞ ⎛ ⎞= − = −⎜ ⎟⎜ ⎟ ⎜− +⎝ ⎠ ⎝⎝ ⎠
∑
∑ ∑ ∑ ∑
∑ ∑ ( )0 0
1 11
mk
k i
m k s ix
m
∞
= =
⎛ + − + − ⎞⎜ ⎟⎟ −⎠⎝ ⎠
∑∑we get [18] Example 1. Count the number of quasicompositions of 4 in 4 parts of order 2 a) From [9] for 2s = we have
( )1 2 1 20 1 2
1 2 1 20 1 2 1 2 1 22
! !! ! ! ! ! !t t k t t k
t t t m
m m mk t t t m t t t t+ = + =
+ + =
⎛ ⎞= =⎜ ⎟ − −⎝ ⎠∑ ∑
( )
( ) ( ) ( )
1 2
1 2
1 21 2 4 1 2 1 22
1 2
1 4 2 0 4; 4, 04 4! 1 2 2 1 4; 2, 14 4 ! ! !
1 0 2 2 4; 0, 24! 4! 4! 19
4 4 0 !4!0! 4 2 1 !2!1! 4 0 2 !0!2!
t t
t tt t
t t t tt t+ =
⋅ + ⋅ = = =⎛ ⎞
= = ⋅ + ⋅ = = = =⎜ ⎟ − −⎝ ⎠ ⋅ + ⋅ = = =
= + + =− − − − − −
∑
b) from [16],[17] and [18] for 2s = we have respectively
02
m
i
m m ik i k i=
⎛ ⎞ ⎛ ⎞⎛ ⎞=⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠∑
02 2
m
i
m m m ik i k i=
−⎛ ⎞ ⎛ ⎞⎛ ⎞=⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠∑
( )02
3 11
1
mi
i
m m m k ik i m=
+ − −⎛ ⎞ ⎛ ⎞⎛ ⎞= −⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠∑
A. Dani
238
that give 4
02
4 4 4 2 4 3 4 46 1 4 3 1 1 6 12 1 19
4 4 2 2 3 1 4 0i
ii i=
⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞= = + + = ⋅ + ⋅ + ⋅ = + + =⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠∑
4
02
4 4 4 4 4 4 3 4 21 4 3 6 1 1 12 6 19
4 4 2 0 4 1 2 2 0i
ii i=
−⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞= = + + = + ⋅ + ⋅ = + + =⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟−⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠∑
( )4
02
4 4 7 3 4 7 4 41 35 16 19
4 3 0 3 1 3i
i
ii=
−⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞= − = − = − =⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟
⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠∑
as is shown above counting with formula [18] is more effective. In continuation we give the list of all 19 quasicompositions of number 4 in 4 parts of order 2
( ) ( ) ( ) ( ) ( ) ( )( ) ( ) ( ) ( ) ( ) ( ) ( )( ) ( ) ( ) ( ) ( ) ( )
0,0,2,2 , 0, 2,0, 2 , 0, 2, 2,0 , 2,0,0,2 , 2,0,2,0 , 2,2,0,0
0,1,1, 2 , 0,1, 2,1 , 0, 2,1,1 , 1,0,1,2 , 1,0,2,1 , 1,1,0,2 , 1,1,2,0
1,2,0,1 , 1, 2,1,0 , 2,0,1,1 , 2,1,0,1 , 2,1,1,0 , 1,1,1,1
Example 2. From Pascal triangle of order 3 we can see that in meting point of row 3 with column 5 is placed number 12, that means number of quasicompositions of 5 in 3 parts of order 3 is 12. That number also we can get from [18]
( )3
03
3 3 7 4 3 7 3 31 21 9 12
5 2 0 2 1 2i
i
ii=
−⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞= − = − = − =⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟
⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠∑
All 12 quasicompositions of 5 in 3 parts of order 3 are ( ) ( ) ( ) ( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )
0,2,3 , 0,3,2 , 2,0,3 , 2,3,0 , 3,0,2 , 3,2,0
1,1,3 , 1,3,1 , 3,1,1
1,2,2 , 2,1, 2 , 2, 2,1
LITERATURE
[1] Сачков В.Н., Введение в комбинаторньiе методьi дискретной матема-тики, Мockвa 1982 (384 p.) [2] I.P.Goulden D.M.Jackson, Combinatorial enumeration, ( пeрeчиcлиteльhaя komбиhatopиka) in russian Mockвa 1990 (504 p.) [3] С.В. Яaблoнсkий, Введение в дискретную математику, Mockвa 1986 (384 p.) [4] Б.А.Бoндаpeнko, Обoбщehыe tpeугoльники и пиpamиди пackaля, Тaшkeнt 1990 (192 p.) [5] B.И.Бapaнob Б.C.Cтeчkин, Эkctpemaльhыe komбиhatophыe зaдaчи и иx пpилoжehия, Мockвa 1989 (160 p.) [6] D. Cvetković, Diskretne matematičke strukture,,beograd 1978 (152 p.) [7] H. Wilff, Generatingfunctionology, Philadelphia 1994 (226 p.)
1Secondary school ,,Niko Nestor,, 6330 Struga, Macedonia e-mail: [email protected]