Quantum Games: Mixed Strategy Nash’s

Equilibrium Represents Minimum Entropy

Edward Jiménez
Experimental Economics, Todo1 Services Inc, Miami Fl 33126
Research and Development Department, Petroecuador, Quito Ecuador, 5932
Tel. 59322291999; Fax 59322524766 (ejimenezecu@yahoo.com) and Post Office Box
Av 6 de Diciembre y Paul Rivet (Edif El Pinar) Quito- Ecuador.

November 11, 2002

Abstract
This paper introduces Hermite’s polynomials, in the description of quan-
tum games. Hermite’s polynomials are associated to the gaussian probabil-
ity density . The gaussian probability density represents (minimum disper-
sion ) and permit us to analyze the convergence to the Nash’s equilibrium
(minimun entropy ME or maximum utility MU). The ME concept is re-
lated to Quantum Games. Some questions arise after carrying out this
exercise: i) What does the Heisemberg’s uncertainly principle represent in
Game Theory and Time Series?, and ii) What does the postulates of the
Quantum Mechanics indicate in Game Theory and Economics?.

Journal of Economics ClassiÞcation C62; C70; C73; C79

Keywords: Quantum Games, Minimum Entropy, Time Series, Nash-Hayeck

equilibrium.

1 INTRODUCTION
The quantum games and the quantum computer are deeply related. Quantum
Computer is one of the modern paradigms in Computer Science and Game Theory
[10], [9], [8], [26], [44] [45], [58]. Quantum Computer will allow to increase the
processing speed and the effective search in the databases. The quantum games
allow to incorporate the algorithms of Quantum Theory to Game Theory [8], this
simple extrapolation allows to solve the Prisoner’s Dilemma and demonstrate that
the cooperative equilibrium is viable and stable with probability different to zero.

1
 “At the most abstract level, game theory is about numbers of entities that
are efficiently acting to maximize or minimize[10]. For a quantum physicist, it is
then legitimate to ask: what happens if linear superpositions of these actions are
allowed for?, that is, if games are generalized into the quantum domain. For a
particular case of the Prisoner’s Dilemma, we show that this game ceases to pose
a dilemma if quantum strategies are implemented for.” They show that classical
strategies are particular cases of quantum strategies.
Eisert. Wilkens, and Lewenstein 2001, not only give a physical model of
quantum strategies but also gives the idea of identifying moves with quantum op-
erations and quantum properties. This approach appears to be fruitful in at least
two ways. By the Þrst hand several recently proposed applications of quantum
information theory, can already be conceived as competitive situations, where
several parts with more or less opposed motives to interact. These parts may
apply quantum operations on a bipartite quantum system. On the other hand a
generalization of the theory of decisions in the domain of quantum probabilities
seems interesting, as the roots of game theory are partly in probability theory.
In this context it is of interest to investigate what solutions are attainable if su-
perpositions of strategies are allowed. A game is also related to transference of
information. It is possible to ask, what happens if these carriers of information
are taken to be quantum systems?, quantum information being a fundamental
notion of information. The Nash’s equilibrium concept in quantum games is es-
sentially the same that in game theory but one difference is that the strategies
are in function of quantum properties of the physical system.
This paper essentially explicits the existent relationship among Quantum Me-
chanics, Nash’s equilibria, Heisenberg Uncertainty Principle, Minimum Entropy
and the Time Series. The Heisenberg’s uncertainty principle is one of the angular
stones in Quantum Mechanics. One application of the uncertainty principle in
Time Series is related to Spectral Analysis “The more precisely the aleatory
variable VALUE is determined, the less precisely the frequency VALUE is known
in this instant, and conversely”. This principle indicates that the product of the
standard deviation of one aleatory variable xt by the standard deviation of its
frequency w is greater or equal to 1/2.
This paper is organized as follows:
Section 2: Quantum Games and Hermite’s Polynomials, Section 3: Time Se-
ries and Heisenberg’s Principle, Section 4: Quantum Games Application, Section
5: Application of Heisenberg’s Principle in Time Series, Section 6: Conclusion
and Section 7: Appendix

2 Quantum Games and Hermite’s Polynomials

Let Γ = (K, S, v) be a game to n−players, with K the set of players k = 1, ..., n.
The Þnite set Sk of cardinality lk ∈ N is the set of pure strategies of each player

2
where k ∈ K, skji ∈ Sk , jk = 1, ..., lk and S = ΠK Sk set of pure strategy proÞles
with s ∈ S an element of that set, l = l1 , l2 , ..., ln represent the cardinality of S
. The vectorial function v : S → Rn associates every proÞle s ∈ S the vector of
utilities v(s) = (v 1 (s), ..., vn (s))T , where v k (s) designates the utility of the player
k facing the proÞle s. In order to get facility of calculus we write the function
v k (s) in one explicit way vk (s) = v k (j1 , j2 , ..., jn ).The matrix vn,l represents all
points of the cartesian product Πk∈K Sk . The vector vk (s) is the k− column of v.
If the mixed strategies are allowed then we have:
( lk
)
X
∆(Sk ) = pk ∈ Rlk : pkjk = 1
jk =1

the unit simplex of the mixed strategies of player k ∈ K, and pk = (pkjk ) the
probability vector. The set of proÞles in mixed strategies is the polyhedron ∆
with ∆ = Πk∈K ∆(Sk ), where p = (p1j1 , p2j2 ..., pnjn ), and pk = (pk1 , pk2 , ..., pkln )T .
Using the Kronecker product ⊗ it is possible to write1 :

p = p1 ⊗p2 ⊗... ⊗ pk−1 ⊗pk ⊗pk+1 ⊗ ... ⊗ pn

p(−k) = p1 ⊗p2 ⊗... ⊗ pk−1 ⊗lk ⊗pk+1 ⊗ ... ⊗ pn
£ ¤
lk = (1, 1, ..., 1)T , lk l ,1
k T
£ k ¤k
o = (0, 0, ..., 0) , o l ,1
k

The n− dimensional function u : ∆ → Rn associates with every proÞle in

³ ´T
mixed strategies the vector of expected utilities u(p) = u1 (p, v(s)) , ..., un (p, v(s)) ,
where uk (p, v(s)) is the expected utility of the player k. Every ukjk = ukjk (p(−k) , v(s))
represents the expected utility for each player’s strategy and the vector uk is noted
uk = (uk1 , uk2 , ..., ukn )T .

lk
X
uk = ukjk (p(−k) , v(s))pkji
jk =1
0
u = vp
¡ ¢
uk = lk ⊗ vk p(−k)

The triplet (K, ∆, u(p)) designates the extension of the game Γ to the mixed
strategies. We get Nash’s equilibrium
¡ ¢∗ (maximization of utility) if and only if, ∀k,
p, the inequality uk (p∗ ) ≥ uk ( pk , p(−k) ) is respected .
Another way to calculate the Nash’s equilibrium [52], is equaling the values
of the prospective utilities of each strategy when it is possible.
1
We use bolt in order to represent vector or matrix.

3
 ¡ ¢ ¡ ¢ ¡ ¢
uk1 p(−k) , v(s) = uk2 p(−k) , v(s) = ... = ukjk p(−k) , v(s)
lk
X
pkjk = 1 ∀k = 1, ..., n
jk =1
lk ³
X ´2
¡ ¢
σ 2k = ukjk p(−k) , v(s) − uk pkjk = 0
jk =1
¡ ¢∗
If the resulting system of equations doesn’t have solution p(−k) then P we
propose the Minimum Entropy Method. This method is expressed as M inp ( k Hk (p)) ,
where σ 2k (p∗ ) standard deviation and Hk (p∗ ) entropy of each player k.

¢∗ ³¡´
σ 2k (p∗ ) ≤ σ 2k
pk , p(−k) or
³¡ ¢∗ ´
∗ k (−k)
Hk (p ) ≤ Hk p , p

2.1 Minimum Entropy Method

Theorem 1 (Minimum Entropy Theorem). The game entropy is minimum only P
in mixed strategy Nash’s equilibrium. The entropy minimization program M inp ( k Hk (p)) ,
is
³ equal
´ to standard deviation minimization program. M inp (Πk σ k (p)) , when
ukjk has gaussian density function or multinomial logit.

In Hayek’s language equilibrium means the order state or minimum entropy.

The order state is opposite to entropy (disorder measure). There are some intel-
lectual inßuences and historical events which inspired Hayek to the development
of the idea of spontaneous order. Here we present the technical tools in order to
study the order state.

Case 1 If the probability density of a variable X is normal: N (µk , σ k ), then its

entropy is minimum for the minimum standard deviation (Hk )min ⇔ (σ k )min .
∀k = 1, ..., n.
R +∞
Proof. Let the entropy function Hk = − −∞ p(x) ln(p(x))dx and p(x) the
normal density function.

Z +∞
Hk = − p(x) ln(p(x))dx
−∞
Z +∞ Ã Ã ! µ ¶2 !
1 x − uk
Hk = − p(x) ln p − √ dx
−∞ 2πσ 2k 2σ k

4
 developing the integral we have

³√ ´ Z +∞
Hk = ln 2πσ k p(x)dx
−∞
Z +∞
1
+ 2 (x − uk )2 p(x)dx
2σ k −∞
³√ ´ µ ¶
σ 2k 1 √
= ln 2πσ k + 2 = + ln 2π + ln σ k
2σ k 2
For a game to n− players the total entropy can be written as continues:

X
n µ¶
1 √
Hk = n + ln 2π + ln(Πnk=1 σ k )
2
k=1
à !
X
M inp Hk (p) ⇔ Minp (Πk σ k (p))
k

after a little of calculus it is possible to demonstrate that

à n !
X
M inp σ k (p) ⇒ M inp (Πk σ k (p))
k=1

The entropy or measure of the disorder is directly proportional to the standard

deviation or measure of the uncertainty. Clausius , the creator of the entropy
notion, presents it as an evolutionary measure and as the characterization of
reversible and irreversible processes [49].

Case 2 If the probability function of ukjk (p(−k) , v(s)) is a multinomial logit of

parameter λ, then its entropy is minimum and its standard deviation is minimum
for λ → ∞.

Proof. Let pkjk the probability for k ∈ K, jk = 1, ..., lk

k −k ) k −k )
eλujk (p eλujk (p
pkjk = Pl λukj (p−k )
=
k
e k Zk
jk =1
³ ´ P k −k
where Zk λ, uk (p−k ) = ljkk =1 eλujk (p ) represents the partition utility func-
tion.
The entropy Hk (pk ), expected utility E{ukjk (p−k )} = uk (p) , and variance
V ar{ukjk (p−k )} will be different for each player k.

5
 lk
X
Hk (pk ) = − pkjk ln(pkjk )
jk =1
lk
X
uk (p) = pkjk ukjk
jk =1
lk
X
© k −k ª
V ar ujk (p ) = pkjk (ukjk − uk )2
jk =1

Using the explicit form of pkjk , we can get the entropy, the expected utility
and the variance:

Hk (pk ) = ln(Zk ) − λuk (p)

∂ ln(Zk )
uk (p) =
∂λ
k
∂u (p)
= V ar{ukjk (p−k )} = σ 2k
∂λ
k
The equation ∂H∂λ k (p )
= −λσ 2k can be obtained using the last seven equations; it
explains that, when the entropy diminishes, the rationality increases.
The rationality increases from an initial value of zero when the entropy is
maximum, and it drops to its minimum value when the rationality spreads to the
inÞnite value: Limλ→∞ Hk (pk (λ)) = min(Hk )
The standard deviation is minimum in the Nash’s equilibria.
If the rationality grows, then the Nash’s equilibria can be reached when the
rationality spreads to the inÞnite: Limλ→∞ σ k (λ) = 0.
For the chain of previous ideas, we can conclude that the entropy diminishes
when the standard deviation diminishes:

(Hk (p) = (Hk (p))min ) ⇔ (σ k (λ) = (σ k )min )

à !
X
M inp Hk (p) ⇔ M inp (Πk σ k (p))
k

after a little of calculus it is possible to demonstrate that

à n !
X
M inp σ k (p) ⇒ M inp (Πk σ k (p))
k=1

6
Remark 1 The entropy Hk for gaussian probability density and multinomial
¡ √ ¢ P k
logit is written as 12 + ln 2π + ln σ k and ln( ljkk =1 eλujk ) − λuk

Case 3 The special case of Minimum Entropy is when σ 2k = 0 and the utility
function value of each strategy ukjk (p(−k) , v(s)) = uk , is the same ∀jk , ∀k.

In the special case of Minimum Entropy when³ σ 2k → ´ 0, the gaussian density

function can be approximated by Dirac’s Delta δ x − uk .
³ ´
The function δ x − uk is called the “Dirac’ function”. The Dirac’s function
is not a function in the usual sense. It represents ½an inÞnitely short,
¾ inÞnitely
∞ if x = uk
strong unit-area impulse. It satisÞes δ (x − uk ) = and can be
0 if x 6= uk
obtained at the limit of functions [7]:
µ ¶2
k
1 − x−u
√
2σk
δ (x − uk ) = Limσk →0 √ e
σ k 2π

Case 4 If we can measure the standard deviation σ 2k 6= 0, then Nash’s equilibrium

represents minimum standard deviation (σ k )min , ∀k ∈ K.

uk1 (p(−k) , v(s)) 6= uk2 (p(−k) , v(s)) 6= ... 6= ukn (p(−k) , v(s))
X lk
pkjk = 1
jk =1
lk
X
σ 2k = (ukj (p(−k) , v(s)) − uk )2 pkjk > 0
jk =1

2
Theorem 2 Minimum Dispersion. The gaussian density function ³ f (x) = ´ |φ(x)|
represents the solution of one differential equation given by 2σx2 + ∂x
∂
φ(x) =
x
0 related to minimum dispersion of a¡ lineal¢ combination between the stochastic
∂
variable x, and a Hermitian operator −i ∂x .

bB
Proof. This proof is extended in [51], [7]. Let A, b be Hermitian operators
2
which do not commute, because we can write,
h i ³ ´
b b b b b
A, B = AB − B A = iCb

2 eb
b have the next property: A eb
b∗ , the transpose operator A
The hermitian operator A =A is
b∗
equal to complex conjugate operator A .

7
 Easily we can veriÞed that the standard deviations satisfy the same rule of
commutation:

h i h i h i
b b b b b b
A, B = 4A, 4B = A − A, B − B
h i
b 4B
4A, b = 4A4
b Bb − 4B4
b A b = iC
b

Let J(α)be the positive function deÞned as expected value hi.

¿³ ³ ´ ³ ´´+ ³ ³ ´ ³ ´´À
J(α) = α 4A b + 4B b α 4A b + 4B b
³ ³ ´ ³ ´´+
By hypothesis the Hermitian operators can be expressed as α 4Ab + 4B
b
³ ³ ´ ³ ´´
b b
= α 4A + 4B . Using this property,
¿³ ´2 À ¿³ ´2 À D E
J(α) = α 2
4Ab + 4Bb + iα 4A4b B b − 4B4
b A
b

the expression J(α) has the minimum value when J 0 (α) = 0, thus the value
of α and Jmin are

D E D E
b b
i 4A4B − 4B4A b b Cb
α = − ¿³ ´2 À = ¿³ ´2 À
2 4Ab 2 4A b
D E D E
b2
C ¿³ ´2 À Cb2
Jmin = ¿³ ´2 À + b
4B − ¿³ ´2 À ≥ 0
4 4A b 2 4A b

simplifying Jmin

¿³ ´2 À ¿³ ´2 À
b b 1 D b2 E
4A 4B ≥ C
4
1
σ Abσ Bb ≥
2
We are interested in knowing the explicit form of the function φ(x)
£ when ¤the
b b ∂ ∂
inequality is transformed into equality and A = x, B = −i ∂x , where x, −i ∂x =
i for which we need to solve the following differential equation.

8
 ³ ´
b + i4B
α4A b φ(x) = 0
 D E 
 Cb 
 ¿³ b + i4B
´2 À 4A b  φ(x) = 0
 
2 4A b
D E D E
b b ∂ b
if A = x, B = −i ∂x , and B = 0 then C b = 1 and
µ ¶
x − hxi ∂
+ φ(x) = 0
2σ 2x ∂x
the solution is

µ ¶1/2
1 (x−hxi)2
φ(x) = √ e− 4σ2
σ x 2π
(x−hxi)2
1 −
f (x) = |φ(x)|2 = √ e 2σ2x
σ x 2π

Remark 2 The gaussian density function f(x) is optimal in the sense of mini-
mum dispersion.
According to the last theorem ukjk = x is a gaussian aleatory variable with
h i
probability density fk (x), ∀k, and for case of calculation E ukjk = uk . We can
write
2
(x−uk )
1 −
2σ 2
fk (x) = √ e k
σ k 2π
Corollary 1 The central limit theorem which establishes that “The mean uk of
random variable ukjk has normal density function”, is respected in Minimum
Entropy Method.

2.2 Quantum Games Elements

A complete relationship exists between Quantum Mechanics and Game Theory,
both share an aleatory nature. Due to this nature it is necessary to speak of
expected values. The Quantum Mechanics deÞnes observed values or expected
values.

9
 Table 1: Quantum Mechanic and Game Theory properties

Quantum Mechanic Game Theory

Particle: k = 1, ..., n Player : k = 1, ..., n
Quantum element Player type
Interaction Interaction
Quantum state: j = 1, ..., lk Strategy: j = 1, ..., lk
Energy e Utility u
Superposition of states Superposition of strategies
State function State function
Probabilistic and optimal nature Probabilistic and optimal nature
Uncertain Principle Minimum Entropy
Matrix operators Matrix operators
Variational Calculus, Optimal Control,
Information Theory Information Theory
Complexity Complexity
variety: number of particles n variety: number of players n
variability: number of interactions n! variability: number of interactions n!
cuantitative: Mathematical model cuantitative: Mathematical model
Observable value: E [e] Observable value: E [u]
The entities communicate efficiently The players interact efficiently
Entropy Minimum Entropy

According to the Quantum Mechanics an uncertainty exists in the simultane-

ous measure of two variables such as: position and momentum, energy and time
[7]. A quantum element that is in mixed state makes it respecting the overlap-
ping probabilistic principle of pure states. For its side in Game Theory a player
that uses a mixed strategy makes it respecting the vNM function or overlapping
probabilistic of pure strategies [6]. The state function in quantum mechanics
doesn’t have equivalent in Game Theory, for that we will use state function for
analogy with the Quantum Mechanics. The comparison between Game Theory
and Quantum Mechanics can turn in an explicit way in Table 1. If the deÞnition
of rationality in Game Theory represents a optimization process, we can say that
quantum processes are essentially optimal, therefore the nature of both processes
is similar

Proposition 1 Let k = 1, ..., n be players with lk strategies for every one jk =

mk1 , ..., mklk . The utility µk , converges to the Nash’s equilibria. and follows a nor-
µ
− 1 ( u − σk )2
h i
e 2 σ√ k k
mal probability density ρk (u) = σ 2π
. The expected utility is E ukjk = µk .
k

10
Definition 1 (Tactics of Sub- Strategies) For a correct description of the strategy
evolution jk , it is necessary the introduction of sub strategies mkjk ∈ Mjkk , ∀k, jk
¯ ¯
¯ k ¯2
with a probability given by ¯¯bmk ¯¯ where mkjk ∈ M k . Let Mjkk be the set of sub-
jk

strategies with ∪ljkk =1 Mjkk = M k , M k = {0, 1, ..., ∞} and Mjkk ∩Mjkl = ∅ when k 6=
l. A strategy jk is constituted of several sub strategies mkjk . With the introduction
of sub strategies the¯ probability ¯ of each strategy is described in the following way
¯ k ¯2 P ¯ k ¯2
¯bj ¯ = ¯ ¯
k mkj ∈Mjk ¯bmkj ¯ and we don’t need any approximation of the function
P∞ k k k
P∞
mkj =0 cmkj (t)ϕmkj (x, λ), where ψ(x, λ) = mk =0 cmk (t)ϕmkj (x, λ), cmkj (t) =
k
¯ k ¯2 ¯k ¯2 ¯ jk ¯2 jk k k
¯ ¯ ¯ ¯ P ¯ ¯
bkjk eiwk t ,¯¯c2mk ¯¯ = ¯¯bkmk ¯¯ = pkmk and ∞ ¯ k ¯
mk =0 ¯bmk ¯ = 1, ∀k.
jk jk jk jk jk

Theorem 3 The normal probability density ρ(u) = |ψ (x, λ)|2 ¯can be obtained
¯2
using Hermite orthogonal polynomials Hk (x). The parameter ¯bkjk ¯ , indicates
the probability value of playing jk strategy. The state function ϕk (x, λ) measure
the behavior of jk strategy (jk one level in Quantum Mechanics).
P∞ The dynamic
behavior of jk strategy can be written as ψ(x, λ) = mkj =0 cmkjk (t)ϕmkj (x, λ).
P k k
One approximation is ψ(x, λ) = mk ∈Z + cmkj (t)ϕmkj (x, λ)
jk k k

The function ψ(x, λ) is determined as follows:

1 −( √x −λk )2
ψ(x, λ) = p √ e 2
σ k 2π
∞
X mk µ ¶
1 (λk ) jk − x2 x
= p √ e 2 Hmkj √
k σ k 2π mkjk ! k 2
m =0jk

√u , µ
where: x = 2σk
λ= 2σk
, and

D E
ϕmkj , ϕmkj = 1,
D k k
E
ϕmkj , ϕmkj = 0, ∀mkjk 6= mkjl
k l
2
³ ´
− x2
e Hmkj √x2
ϕmkj (x, λ) = q √
k

k mk √
mkjk ! 2 jk π

Proof. The Hermite polynomials properties will be necessary to express ρ(u)

in function of Hk (x).
Using generating function ψ(x, λk ) we can write

11
 1 x 2
p √ e−( 2 −λk )
√
(Q1)
σ k 2π
à k
! µ ¶
∞
X (λk )mjk 2 x
− x2
= p √ e Hmkj √
mk =0
mk
jk ! σ k 2π k 2
jk

Let ψ(x, λk ) be the generating function

∞
X
ψ(x, λk ) = bmkj ϕmkj (x, λk ) (Q2)
k k
mkj =0
k
P∞
In order to get ψ(x, λ) = mkj =0 bmkjk ϕmkj (x, λ) with the equations (QG1,
k k
PH1, PH2) and:
¿ µ ¶ µ ¶À
− x2
2 x − x2
2 x √ mk √
e Hmkj √ , e Hmkj √ = mkjk ! 2 jk π
k 2 k 2
we can write:

1 −
(x−λk )2
p √ e 2 (Q3)
σ k 2π
 q ³ ´
k √ mkj −1   e− x22 H k √x 
X∞ m
 (λk )q 2 k 
jk mj 2
= q k 
k √ √ m k √
mkj =0 mjk ! σ n mkj ! 2 jk π
k k

Comparing (Q2) and (Q3) is easy to conclude

q
√ mkj −1
mkj
(λk ) k 2 k
bmkj = q √ (Q4)
k
mkjk ! σ k
x2
³ ´
e− 2 Hmkj √x2
ϕmkj (x, λk ) = q √
k
(Q5)
k mk √
mkjk ! 2 jk π

¯ ¯2
¯ ¯
The ¯bmkj ¯ parameter represents the probability of mkjk sub strategy of each
k
k player. The utility of vNM can be written as:

12
 h i lk
X ¯ k ¯2 k
k k −k
E ujk = u (p ) = ¯bj ¯ uj (Q5.1)
k k
jk =1
¯ k ¯2 X ¯ ¯2
¯b ¯ = ¯ k ¯
jk ¯bmkj ¯ (Q5.2)
k
mkj ∈Mjk
k k

Using the operators a, a+ let us write the equations:

∂
(x + ∂x )
aϕmkj (x, λk ) = ¡√ ¢ 3/2
ϕmkj (x, λk ) (Q6)
k k
2
√
aϕmkj (x, λk ) = nϕmkj −1 (x, λk ) (Q6.1)
k k

∂
(x+ ∂x )
The equation (Q6.1) indicates that the operator a = √ 3/2 decrease de level
( )
2
of k player. The operator a can decrease the sub strategy description mkjk →
mkjk − 1.

∂
¡¢
+ x − ∂x
a ϕmkj (x, λ) = ¡√ ¢3/2 ϕmkj (x, λ) (Q7)
k k
2
√
a+ ϕmkj (x, λ) = n + 1ϕmkj +1 (x, λ) (Q7.1)
k k

∂
(x− ∂x )
The equation (Q7.1) indicates that the operator a+ = √ 3/2 increase the
( 2)
level of k player. The operator a+ can increase the sub strategy description
mkjk → mkjk + 1.
The Hermite polynomials form an orthogonal base and they possess desir-
able properties that permit to express a normal probability density as a linear
combination of them.

2.3 Hermite’s Polynomials Properties

The next properties are taken of [7], [51], [35].
Hermite polynomials obey the next differential equation:
00 0
Hmk (x) − 2xHmk (x) + 2mkjk Hmkj (x) = 0, mkjk = 0, ..., ∞ (PH0)
jk jk k

The Principal Solution of Hermite differential equation3 is

3
H0 (x) = 1, H1 (x) = 2x,
H2 (x) = 4x2 − 2, H3 (x) = 8x3 − 12x

13
 mkj 2
x2 d
k e−x
Hmkj (x) = (−1)e k (PH1)
k
dxmjk
k mkj (mkjk − 1) k
Hmkj (x) = (2x)mjk − k (2x)mjk −2
k 1!
mkjk (mkjk − 1)(mkjk − 2)(mkjk − 3) k
+ (2x)mjk −4 +
2!
Taking the explicit value of mkjk −derivative of (PH1):

mkj
d k Hmkj (x) k

mkj
k
= mkjk !2mjk (PH2)
dx k
mk
d jk Hmkj ( √x2 ) √ mk
k
mkj
= mkjk ! 2 jk (PH2.1)
dx k

Decreasing operator:

dHmkj (x)
k
= 2nHmkj −1 (x) (PH3)
dx k

Increasing operator:
d
2x(x − )H k (x) = Hmkj +1 (x) (PH4)
dx mjk k

Recurrence formula:

Hmkj +1 (x) = 2xHmkj (x) − 2mkjk Hmkj −1 (x) (PH5)

k k k

Normalization condition:

Z +∞
2
e−x Hm
2
k (x)dx (PH6)
jk
−∞
Z +∞
k
dmjk Hmkj (x) k √
−x2
= e mkj
k
dx = mkjk !2mjk π
−∞ dx k
The next condition expresses the orthogonal property of Hermite’s polyno-
mials:

D x2 2 E
− 2 − x2
e Hmkj (x), e Hmkj (x)
k k
Z +∞
2
= e−x Hmkj (x)Hmkj (x)dx = 0 ∀mkjk 6= mkjl
k l
−∞

14
 Generating Function is obtained using the Taylor development.

X (λk )mkjk
x2 −(x−λk )2
F (x, λk ) = e e = Hmkj (x) (PH7)
mkjk ! k
mkj
k

Quantum Games generalize the representation of the probability in the utility

function of vNM (von Newmann and Morgenstern). For understanding Quantum
Games it is necessary to use properties of the quantum operators in harmonic os-
cillator and the properties of Hilbert space. The harmonic oscillator is presented
as the paradigm in Quantum Games.

2.4 Quantum Games Properties

According to Quantum Mechanics[7], [51], [35], a quantum system follows the

next postulates or axioms:

Axiom 1 At a Þxed time to , the state of a physical system is deÞned by specifying

a ket |ψ(to )i beloging to the state space (Hilbert space) V.

Axiom 2 Every measurable physical quantity Λ is described by the operator A

acting in V ; this operator is an observable.

Axiom 3 The only possible result of the measurement of a physical quantity Λ

is one of the eigenvalues of the corresponding observable A.

Axiom 4 When Pa physical

¯ ® quantity Λ is measured on a system in the normalized
state |ψ(t)i = cj ¯ϕj , the probability P (bj ) of obtaining the non-degenerate
¯­ ®¯2 ¯ ®
eigenvalue cj of the corresponding observable A is: P (cj ) = ¯ ϕj , ψ ¯ where ¯ϕj
is the normalized eigen vector of A associated with the eigen value cj .

Let n be players with k = 1, ..., n , lk strategies and mkjk ∈ Mjkk of each player
k.
Superposition Principle:
∞
X D E
k
ψ = ckmk ϕmkj , ϕmkj , ϕmkj = δ mkj mkj (QG1)
j k k k l k l
mkj =0
k

Time Evolution:
√
ckmk = bkmk eiwk t , i= −1 (QG2)
jk jk

Inner Product:

15
 D E
ckmk = ψ k , ϕmkj (QG3)
jk k

Quantum Property

¯ ¯ ¯ ¯2
¯ k ¯2 ¯ ¯
¯cmk ¯ = ¯bkmk ¯ = pkmk (QG4)
jk j j k k
∞
X lk
X
pkmk = pkj = 1
j k k
mkj =0 jk =1
k
X
pkj = pkmk
k jk
mkj ∈Mjk
k k

Mean Value
­ ®
E [xk ] = xk = ψ k |x| ψ k (QG5)
Normality Condition
­ ® ¯ ¯2
ρk = ψ k , ψ k = ¯ψ k ¯ (QG6)

3 Time Series and Uncertainty Principle of Heisen-

berg
Remember some basic deÞnitions about probability and stochastic calculus.
A probability space is a triple (Ω, =, P ) consist of:

• A set Ω that represents the set of all possible outcomes of a certain random
experiment.

• A family = of subsets of Ω with structure of σ-algebra:

— ∅∈=
— A ∈ = =⇒ Ac ∈ =
— A1 , A2 , ... ∈ = =⇒ ∪∞
i=1 Ai ∈ =

• A function P : Ω → [0, 1] such that:

— P (∅) = 0, P (Ω) = 1

16
 — If A1 , A2 , ... ∈ = form a Þnite or countably inÞnite collection of dis-
joints sets (that is, Ai ∩ Aj = ∅ if i 6= j) then

∞
X
P (∪∞
i=1 Ai ) = P (Ai )
i=1

The elements of the σ-algebra = are called events and P is called a probability
measure.

A random variable is a function:

X : Ω→R
ω → X(ω)

=-measurable, that is, X −1 (B) ∈ = for all B in the Borel’s σ-álgebra of R,

noted ß(R).
If X is a random variable on (Ω, =, P ) the probability measure induced by
X is the probability measure PX (called law or distribution of X) on ß(R) given
by

PX (B) = P {ω : X(ω) ∈ B} , B ∈ ß(R)

The numbers PX (B), B ∈ß(R), completely characterize the random variable
X in the sense that they provide the probabilities of all events involving X.

Definition 2 The distribution function of a random variable X is the function

F = FX from R to [0, 1] given by

F (x) = P {ω : X(ω) ≤ x} x∈R

Since, for a < b, F (b) − F (a) = P {ω : a < X(ω) ≤ b} = PX (a, b], F is a
distribution function corresponding to the Lebesgue Stieltjes measure PX . Thus
among all distribution functions corresponding to the Lebesgue-Stieltges measure
PX , we choose the one with F (∞) = 1, F (−∞) = 0. In fact we can always supply
the probability space in a canonical way; take Ω = R, = =ß(R), with P the
Lebesgue-Stieltges measure corresponding to F.

Definition 3 A stochastic process X = {X(t), t ∈ T } is a collection of random

variables on a common probability space (Ω, =, P ) indexed by a parameter t ∈
T ⊂ R, which we usually interpret as time [33]. It can thus be formulated as a
function X : T × Ω → R.

17
 The variable X(t, •) is =−measurable in ω ∈ Ω for each t ∈ T ; henceforth
we shall often follow established convention and write Xt for X(t). When T is
a countable set, the stochastic process is really a sequence of random variables
Xt1 , Xt2 , ..., Xtn , ..

Definition 4 If a random variable X = X(t) has density ρ(x) and f : X → Y

ρ(f −1 (y))
is one-to-one, then y = f(x) has density g(y) given by g(y) = f 0 (f −1 (y)) =
ρ (f −1 (y)) (f −1 )0 (y) [14] ( chapter 6). The densities ρ(x) and g(y) shows that
“If X is a random variable and exists f −1 (y) then Y is a random variable and
conversely”

Definition 5 Inversion Formula: If h is a characteristic function of the bounded

distribution function F, and F (a, b] = F (b) − F (a) then

Z +c
1 e−iwa − e−iwb
F (a, b] = lim h(w)dw
c→∞ 2π −c iu
for all points a, b (a < b) at which F is continuous. If in addition, h is Lebesgue
integrable on (−∞, ∞) , then the function f given by
Z +∞
1
f (x) = e−iwx h(w)dw
2π −∞
R +x
is a density of F, that is, f is nonnegative and F (x) = −∞ f (t)dt for all
x; furthermore, F 0 = f everywhere. Thus in this case, f and h are “Fourier
transform pairs”:

Z +∞
1
h(w) = z(f (x)) = eiwx f(x)dx
2π −∞
Z +∞
1
f(x) = z−1 (h(w)) = e−iwx h(w)dw
2π −∞

Definition 6 If X is an aleatory variable., f a Borel measurable function [on (<, <)] ,

then f(X) is an aleatory variable[28]. If W is r.v., h(W ) a Borel measurable
function [on (<, <)] , then h(W ) is r.v.

Theorem 4 The Possibility Theorem: σ x σ w ≥ 12 . “The more precisely the

aleatory variable VALUE x is determined, the less precisely the frequency VALUE
w is known in this instant, and vice versa”. Let X = {X(t), t ∈ T } be the stochas-
tic process with density function f (x) and the random variable w with character-
istic function h(w). If the density functions can be written as f (x) = ψ(x)ψ(x)∗
and h(w) = φ(w)φ(w)∗ , then: σ 2x σ 2w ≥ 14 .

18
 If the functions ψ(x = ±∞) = 0 and φ(w = ±∞) = 0 then f (x = ±∞) = 0
and h(w = ±∞) = 0.
Proof. The next proof is based in [13].
Let the function ψ(x) deÞned as,

Z +∞
−1 1
ψ(x) = z [φ(w)] = √ φ(w)e−iwx dw ∀x, ψ(x = ±∞) = 0 (Q0)
2π −∞

Here w is the frequency, different to ω ∈ Ω .

Then φ(w) is

Z +∞
1
φ(w) = z [ψ(x)] = √ ψ(x)eiwx dx ∀x, φ(w = ±∞) = 0 (Q1)
2π −∞

According to Fourier’s transform properties, we have:

Z +∞
∂ n φ(w) in
n
= √ xn ψ(x)eiwx dx = in z [xn ψ(x)] (Q2)
∂w 2π −∞
Z
n
∂ ψ(x) (−i)n +∞ n
n
= √ w φ(w)e−iwx dw = (−i)n −1 z [wn φ(w)]
∂x 2π −∞
where:

µ ¶ Z +∞ n
∂ n ψ(x) 1 ∂ ψ(x) iwx
z = √ e dx = (iw)n φ(w) (Q3)
∂xn 2π −∞ ∂x n
µ n ¶ Z +∞ n
∂ φ(w) 1 ∂ φ(w) −iwx
z−1
n
= √ n
e dw = (−ix)n ψ(x)
∂x 2π −∞ ∂x
By Parseval Plancherel theorem
Z +∞ Z +∞
∗
φ(w)φ(w) dw = ψ(x)ψ(x)∗ dw (Q4)
−∞ −∞

Using Schwartz’s inequality

¯Z ¯2 Z Z
¯ +∞ ¯ +∞ +∞
¯ G Hdx¯¯ ≤
∗ 2
|G| dx ∗ |H|2 dx (Q5)
¯
−∞ −∞ −∞
R +∞ R +∞
and the probability properties −∞
f (t)dt = 1, −∞
h(w)dw = 1, it is evident
that:

19
 Z +∞
φ(w)φ(w)∗ dw = 1 (Q6)
−∞
Z +∞
ψ(x)ψ(x)∗ dw = 1
−∞

Let σ 2w
be the standard deviation of frequency w, and σ 2x the standard devi-
ation of the aleatory variable x.
Z +∞
σ 2w = w 2 φ(w)φ(w)∗ dw (Q7.1)
−∞
Z +∞
σ 2x = x2 ψ(x)ψ(x)∗ dx (Q7.2)
−∞
with Parseval Plancherel theorem
Z +∞ Z +∞
∂ψ(x) ∂ψ(x)∗
dx = (wφ(w)) (wφ(w))∗ dw (Q8.1)
−∞ ∂x ∂x −∞
Z +∞ Z +∞
ψ 0 ψ 0∗ dx = w2 φ(w)φ(w)∗ dw (Q8.2)
−∞ −∞

Making the product (Q7.1)×(Q7.2) and using Q8, it follows that

Z +∞ Z +∞
σ 2w σ 2x = 2
w φ(w)φ(w) dw ∗
x2 ψ(x)ψ(x)∗ dx (Q9.)
−∞ −∞
Z +∞ Z +∞
= ψ 0 ψ 0∗ dx x2 ψ(x)ψ(x)∗ dx (Q9.2)
−∞ −∞
0
Let G = xψ and H = ψ in Q5 shows that the right side of (Q9.2) exceeds a
certain quantity A
Z +∞ Z +∞ ¯Z +∞ ¯2
¯ ¯
2
|xψ| dx ∗ 0 2 ¯
|ψ | dx ≥ ¯ xψ ψ dx¯¯
∗ 0
(Q10)
−∞ −∞ −∞
R +∞ 0 0∗ R +∞ 2
We will note σ w σ x = −∞ ψ ψ dx −∞ x ψ(x)ψ(x)∗ dx ≥ |A|2 . If the variable
2 2
R +∞ R +∞
A is given by A = −∞ xψ ∗ ψ 0 dx then |A|2 = −∞ x2 dx |ψψ ∗ | |ψ 0 ψ 0∗ | . By deÞnition
0 ∗
F (x) = f = ψψ

f 0 = ψ 0 ψ ∗ + ψψ 0∗
Z +∞ Z +∞
∗ 0 ∗ 0∗
A+A = xdx (ψ ψ + ψψ ) = xf 0 (x)dx
−∞ −∞
Z +∞
A + A∗ = − f (x)dx = −1 (Q11)
−∞

20
 Notice that A + A∗ = −1 and using the complex number property 2 Re (A) =
−1 or 2 |A| cos θ = −1 where θ phase of A, we can write 4 |A|2 = cos12 θ . Since
−1 ≤ cos θ ≤ 1 then |A|2 ≥ 14 and therefore:
1
σ 2x σ 2w ≥ (Q12)
4

x 2 2 2
w σ
Example 1 If f (x) = √1 e− 2σ2 then g(w) = √σ e− 2
σ 2π 2π

Lemma 1 Let a time series xt with a spectrum of frequencies wj where each

frequency is an aleatory variable. This spectrum of frequencies can be obtained
with a minimum error (standard deviation of frequency). This minimum error
σ w for certain frequency w is given by the following equation σ w min = 2σ1x . The
expected value E [wj ] of each frequency can be determined experimentally.

An occurrence of one particular frequency wj is deÞned in an conÞdence

interval given by
· ¸
1 1
wj ∈ E [wj ] − , E [wj ] + (Q13)
2σ x 2σ x
This last lemma agrees and supplements the statement of [21], “We Should
interpret 12 (a2j + b2j ) not as the portion of the variance of X that is due to cycles
with frequency exactly equal to wj , but rather as the portion of the variance of
X that is due to cycles with frequency near of wj , ” where:

M
X
xt = µ + {aj cos(wj (t − 1)) + bj sin(wj (t − 1))} + ut or (Q14)
j=1
xt = c + φ1 xt−1 + ... + φp xt−p + εt + θ1 εt−1 + ... + θq εt−q

The only condition that imposes the Possibility Theorem is exhaustive and
exclusive σ xσ w ≥ 12 , because it includes the Minimum Entropy Theorem or
equilibrium state theorem σ x σ w = 12 . It is necessary to analyze the different
cases of the time series xt using the Possibility Theorem. A time series xt evolves
in the dynamic equilibrium if and only if σ x σ w = 12 . A time series evolves out of
the dynamic equilibrium if and only if σ x σ w > 12 .

21
 Table 2: Evolution of Time Series, out of the equilibrium σ x σ w > 12 .

In this table we can see the different changes in σ x and σ w .

Cases σx σw σxσw Entropy

1 ↑ ↑ > 12 ↑
2 ↑ ↓ > 12 ↑
3 ↓ ↑ > 12 ↑
4 ↓ ↓ > 12 ↑
5 0 ∞ 0 ≯ 12 trivial

Table 3: Evolution of Time Series, in the equilibrium σ x σ w = 12 .

In this table we can see the different changes in σ x and σ w .

Cases σx σw σxσw Entropy

1 ↑ ↓ = 12 ↑
2 ↓ ↑ = 12 ↓
3 max min = 12 min
4 min max = 12 max
5 0 ∞ 0 6= 12 trivial

3.1 Hirschman form of the uncertainty principle

The fundamental analytical effect of Heisenberg principle is that the probability
densities for x and w cannot both be arbitrarily narrow.

Z +∞
H(W ) = dwp(w) log(p(w)) (Q15)
−∞
Z +∞
H(X) = dxp(x) log(p(x))
−∞
µ ¶
1
H(W ) + H(X) ≥ 1 + log
2

Example:
When ψ(x) and φ(w) are Gaussian H(W ) = B + Log(σ W ) and H(X) =
C + Log(σ X ),Hirschman’s inequality become Heisenberg, then inequalities are
transformed in equalities and the minimum uncertainly is minimum entropy. “in
Quantum Mechanic minimum uncertainty product” also obeys a minimum en-
tropy sum.

22
4 Applications of the Models
4.1 Hermites’s Polynomials Application
Let Γ = (K, S, v) be a game to 3−player, with K the set of players k = 1, 2, 3.
The Þnite set Sk of cardinality lk ∈ N is the set of pure strategies of each player
where k ∈ K, skjk ∈ Sk , jk = 1, 2, 3 and S = Πk Sk set of pure strategy proÞles
with s ∈ S an element of that set, l = 3 ∗ 3 ∗ 3 = 27 represent the cardinality of S
. The vectorial function v : S → R3 associates every proÞle s ∈ S the vector of
utilities v(s) = (v1 (s), ..., v 3 (s)), where v k (s) designates the utility of the player
k facing the proÞle s. In order to get facility of calculus we write the function
v k (s) in one explicit way v k (s) = v k (j1 , j2 , ..., jn ).The matrix v3,27 represents all
points of the Cartesian product ΠK Sk . The vector vk (s) is the k- column of v.
The graphic representation of the 3-players game (Figure 1).

Figure: 1 3-players game strategies

In these games we obtain the Nash’s equilibria in pure strategy (Maximum

Utility MU, Table 4) and mixed strategy (Minimum Entropy Theorem MET,
Table 5). After Þnding the equilibria we carried out a comparison with the
obtained results of applying the theory of quantum games developed previously.

23
 Table 4: Maximum Utility maxp (u1 + u2 + u3 )

3- Players Game : Random Utilities

Nash Utilities and Standard Deviations

1 2 3
Max u +u +u = 20,369

1 2 3
u u u
8,7405 9,9284 1,6998
σ1 σ2 σ3
6,3509 6,2767 3,8522
H1 H2 H3
4,2021 4,1905 3,7121

Nash Equilibria: Probabilities, Utilities

Player 1 Player 2 Player 3
1 1 1 2 2 2 3 3 3
p1 p 2 p 3 p1 p 2 p 3 p1 p 2 p 3
1 0 0 1 0 0 0 0 1
1 1 1 2 2 2 3 3 3
u 1 u 2 u 3 u 1 u 2 u 3 u 1 u 2 u 3
8,7405 1,1120 -3,8688 9,9284 -1,2871 -0,5630 -1,9587 5,7426 1,6998
1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3
p 1 u 1 p 2 u 2 p 3 u 3 p 1 u 1 p 2 u 2 p 3 u 3 p 1 u 1 p 2 u 2 p 3u 3
8,7405 0,0000 0,0000 9,9284 0,0000 0,0000 0,0000 0,0000 1,6998

Kroneker Products

u (j1,j2,j3)

u2(j1,j2,j3)

u3(j1,j2,j3)
v1(j1,j2,j3)

v2(j1,j2,j3)

v3(j1,j2,j3)

j1 j2 j3 p1j1 p2j2 p3j3 p1j1p2j2 p1j1p3j3 p2j2p3j3

1
1 1 1 2,9282 -1,3534 -1,9587 1,0000 1,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 -1,9587
1 1 2 6,2704 3,2518 5,7426 1,0000 1,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 5,7426
1 1 3 8,7405 9,9284 1,6998 1,0000 1,0000 1,0000 1,0000 1,0000 1,0000 8,7405 9,9284 1,6998
1 2 1 4,1587 6,9687 4,1021 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
1 2 2 3,8214 2,7242 8,6387 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
1 2 3 -3,2109 -1,2871 -4,1140 1,0000 0,0000 1,0000 0,0000 1,0000 0,0000 0,0000 -1,2871 0,0000
1 3 1 3,0200 2,3275 6,8226 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
1 3 2 -2,7397 3,0191 6,6629 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
1 3 3 1,1781 -0,5630 5,3378 1,0000 0,0000 1,0000 0,0000 1,0000 0,0000 0,0000 -0,5630 0,0000
2 1 1 3,2031 -1,5724 -0,9757 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 1 2 1,9478 2,9478 6,7366 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 1 3 1,1120 6,4184 5,0734 0,0000 1,0000 1,0000 0,0000 0,0000 1,0000 1,1120 0,0000 0,0000
2 2 1 5,3695 5,7086 -0,7655 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 2 2 2,4164 1,6853 7,1051 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 2 3 5,2796 2,5158 -4,7264 0,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 3 1 -4,0524 -4,5759 5,8849 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 3 2 3,8126 -1,2267 4,8101 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
2 3 3 -1,4681 10,8633 0,2388 0,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 1 1 -0,4136 -2,6124 4,5470 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 1 2 2,6579 1,7204 0,7272 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 1 3 -3,8688 4,0884 11,2930 0,0000 1,0000 1,0000 0,0000 0,0000 1,0000 -3,8688 0,0000 0,0000
3 2 1 2,1517 4,8284 14,1957 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 2 2 6,8742 -1,8960 7,4744 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 2 3 2,9484 2,1771 0,0130 0,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 3 1 3,9191 -4,1335 7,4357 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 3 2 -3,8252 3,0861 4,5020 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000
3 3 3 3,6409 3,4438 5,4857 0,0000 0,0000 1,0000 0,0000 0,0000 0,0000 0,0000 0,0000 0,0000

24
 Table 5: Minimum Entropy minp (σ 1 σ 2 σ 3 ) ⇔ minp (H1 + H2 + H3 )

3- Players Game : Random Utilities

Nash Utilities and Standard Deviations
Min (σ1σ2σ3) = 2E-07

1 2 3
u u u
1,917 1,7668 4,984
σ1 σ2 σ3
2Ε−07 0,45383 2,111
H1 H2 H3
0,035 1,74694 3,131

Nash Equilibria: Probabilities, Utilities

Player 1 Player 2 Player 3
p 11 p1 2 p1 3 p2 1 p2 2 p2 3 p 31 p 32 p 33
0,2121 0,2175 0,5704 0,1948 0,3837 0,4215 0,2980 0,3773 0,3247
u 11 u1 2 u1 3 u2 1 u2 2 u2 3 u 31 u 32 u 33
1,9173 1,9173 1,9173 2,1350 2,0716 1,3192 6,6969 5,6543 2,6318
p 1 1 u 1 1 p1 2 u1 2 p1 3 u1 3 p2 1 u2 1 p2 2 u2 2 p 2 3 u2 3 p 3 1 u 3 1 p 3 2 u 3 2 p 3 3 u3 3
0,4067 0,4170 1,0936 0,4159 0,7948 0,5561 1,9957 2,1336 0,8544

Kroneker Products u1(j1,j2,j3)

u2(j1,j2,j3)

u3(j1,j2,j3)
v1(j1,j2,j3)

v2(j1,j2,j3)

v3(j1,j2,j3)

j 1 j2 j 3 p1j1 p2j2 p3j3 p1j1p2j2 p1j1p3j3 p2j2p3j3

1 1 1 2,93 -1,35 -1,96 0,2121 0,1948 0,2980 0,0413 0,0632 0,0581 0,1700 -0,0856 -0,0809
1 1 2 6,27 3,25 5,74 0,2121 0,1948 0,3773 0,0413 0,0800 0,0735 0,4609 0,2603 0,2373
1 1 3 8,74 9,93 1,70 0,2121 0,1948 0,3247 0,0413 0,0689 0,0632 0,5528 0,6837 0,0702
1 2 1 4,16 6,97 4,10 0,2121 0,3837 0,2980 0,0814 0,0632 0,1143 0,4755 0,4405 0,3338
1 2 2 3,82 2,72 8,64 0,2121 0,3837 0,3773 0,0814 0,0800 0,1448 0,5532 0,2181 0,7030
1 2 3 -3,21 -1,29 -4,11 0,2121 0,3837 0,3247 0,0814 0,0689 0,1246 -0,3999 -0,0886 -0,3348
1 3 1 3,02 2,33 6,82 0,2121 0,4215 0,2980 0,0894 0,0632 0,1256 0,3794 0,1471 0,6100
1 3 2 -2,74 3,02 6,66 0,2121 0,4215 0,3773 0,0894 0,0800 0,1591 -0,4358 0,2417 0,5958
1 3 3 1,18 -0,56 5,34 0,2121 0,4215 0,3247 0,0894 0,0689 0,1369 0,1612 -0,0388 0,4773
2 1 1 3,20 -1,57 -0,98 0,2175 0,1948 0,2980 0,0424 0,0648 0,0581 0,1859 -0,1019 -0,0413
2 1 2 1,95 2,95 6,74 0,2175 0,1948 0,3773 0,0424 0,0821 0,0735 0,1432 0,2419 0,2854
2 1 3 1,11 6,42 5,07 0,2175 0,1948 0,3247 0,0424 0,0706 0,0632 0,0703 0,4532 0,2150
2 2 1 5,37 5,71 -0,77 0,2175 0,3837 0,2980 0,0834 0,0648 0,1143 0,6139 0,3700 -0,0639
2 2 2 2,42 1,69 7,11 0,2175 0,3837 0,3773 0,0834 0,0821 0,1448 0,3498 0,1383 0,5929
2 2 3 5,28 2,52 -4,73 0,2175 0,3837 0,3247 0,0834 0,0706 0,1246 0,6576 0,1776 -0,3944
2 3 1 -4,05 -4,58 5,88 0,2175 0,4215 0,2980 0,0917 0,0648 0,1256 -0,5091 -0,2966 0,5395
2 3 2 3,81 -1,23 4,81 0,2175 0,4215 0,3773 0,0917 0,0821 0,1591 0,6064 -0,1007 0,4410
2 3 3 -1,47 10,86 0,24 0,2175 0,4215 0,3247 0,0917 0,0706 0,1369 -0,2009 0,7671 0,0219
3 1 1 -0,41 -2,61 4,55 0,5704 0,1948 0,2980 0,1111 0,1700 0,0581 -0,0240 -0,4440 0,5052
3 1 2 2,66 1,72 0,73 0,5704 0,1948 0,3773 0,1111 0,2152 0,0735 0,1954 0,3703 0,0808
3 1 3 -3,87 4,09 11,29 0,5704 0,1948 0,3247 0,1111 0,1852 0,0632 -0,2447 0,7571 1,2548
3 2 1 2,15 4,83 14,20 0,5704 0,3837 0,2980 0,2188 0,1700 0,1143 0,2460 0,8207 3,1065
3 2 2 6,87 -1,90 7,47 0,5704 0,3837 0,3773 0,2188 0,2152 0,1448 0,9952 -0,4081 1,6357
3 2 3 2,95 2,18 0,01 0,5704 0,3837 0,3247 0,2188 0,1852 0,1246 0,3673 0,4032 0,0028
3 3 1 3,92 -4,13 7,44 0,5704 0,4215 0,2980 0,2404 0,1700 0,1256 0,4923 -0,7026 1,7878
3 3 2 -3,83 3,09 4,50 0,5704 0,4215 0,3773 0,2404 0,2152 0,1591 -0,6084 0,6642 1,0824
3 3 3 3,64 3,44 5,49 0,5704 0,4215 0,3247 0,2404 0,1852 0,1369 0,4983 0,6377 1,3190

25
Table 6: Minimum Entropy: Stone-paper-scissor minp (H1 + H2 + H3 )

3- Players Game : Stone-Paper-Sccisor

Nash Utilities and Standard Deviations
1 2 3
Min (σ1σ2σ3) = 0

1 2 3
u u u
0,5500 0,5500 0,5500
1 2 3
σ σ σ
0,0000 0,0000 0,0000
H1 H2 H3
0,0353 0,0353 0,0353

Nash Equilibria: Probabilities, Utilities

Player 1 Player 2 Player 3
1 1 1 2 2 2 3 3 3
p1 p 2 p 3 p1 p2 p 3 p1 p2 p 3
0,3333 0,3333 0,3333 0,3333 0,3333 0,3333 0,3300 0,3300 0,3300
1 1 1 2 2 2 3 3 3
u 1 u 2 u 3 u 1 u 2 u 3 u 1 u 2 u 3
0,5500 0,5500 0,5500 0,5500 0,5500 0,5500 0,5556 0,5556 0,5556
1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3
p 1 u 1 p 2 u 2 p 3 u 3 p 1 u 1 p 2 u 2 p 3 u 3 p 1 u 1 p 2 u 2 p 3u 3
0,1833 0,1833 0,1833 0,1833 0,1833 0,1833 0,1833 0,1833 0,1833

Kroneker Products

u (j1,j2,j3)

u2(j1,j2,j3)

u3(j1,j2,j3)
v1(j1,j2,j3)

v2(j1,j2,j3)

v3(j1,j2,j3)

1 2 3 1 2 1 3 2 3
j1 j2 j3 p p p p j1p p j1p p j2p
1
j1 j2 j3 j2 j3 j3
stone stone stone 0 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,0000
stone stone paper 0 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,1111
stone stone sccisor 1 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,0000
stone paper stone 0 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,0000
stone paper paper 0 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,1111
stone paper sccisor 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
stone sccisor stone 1 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,1111
stone sccisor paper 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
stone sccisor sccisor 1 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,0000
paper stone stone 1 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,0000
paper stone paper 1 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,1111
paper stone sccisor 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
paper paper stone 1 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,0000
paper paper paper 0 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,0000
paper paper sccisor 0 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,1111
paper sccisor stone 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
paper sccisor paper 0 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,0000
paper sccisor sccisor 0 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,1111
sccisor stone stone 0 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,1111
sccisor stone paper 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
sccisor stone sccisor 0 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,1100 0,0000
sccisor paper stone 1 1 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,1111
sccisor paper paper 1 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,0000
sccisor paper sccisor 1 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,0000 0,1111
sccisor sccisor stone 0 0 1 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,1111
sccisor sccisor paper 1 1 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,1100 0,1100 0,0000
sccisor sccisor sccisor 0 0 0 0,3333 0,3333 0,3300 0,1111 0,1100 0,1100 0,0000 0,0000 0,0000

26
4.2 Time Series Application
The relationship between Time Series and Game Theory appears when we apply
the entropy minimization theorem (EMT). This theorem (EMT) is a way to ana-
lyze the Nash-Hayeck equilibrium in mixed strategies. To introduce the elements
the rationality and equilibrium in the domain of time series can be of a big help,
because it will allow us to study the human behavior reßected and registered
in historical data. The main contributions of this complementary focus to time
series has relationship with econophysics and rationality.

• The human behavior, evolves and it is the result of learning. The objective
of learning is the stability and the optimal equilibrium. Due to the above-
mentioned, we can affirm that if learning is optimal then the convergence
to equilibrium is to a bigger speed that when the learning is sub optimal.
To introduce elements of the Nash’s equilibrium in time series, will allow to
evaluate the learning and the convergence to equilibrium through the study
of historical data.

• One of the branches of Physics called Quantum Mechanics was pioneer in

using Heisemberg uncertainty principle. This paper is simply the applica-
tion of this principle in Game Theory and Time Series.

• Econophysics is a newborn branch of the scientiÞc development that looks

for to establish the analogies between Economics and Physics. The estab-
lishment of analogies is a creative way to apply the idea of cooperative
equilibrium. Product of this cooperative equilibrium will take place syn-
ergies between these two sciences. From my point of view, the power of
the physics is the capacity of equilibrium formal treatment in stochastic
dynamic systems. By its side the power of the Economics is the formal
study of the rationality, the cooperative and not cooperative equilibrium.

• Econophysics is the beginning of an uniÞcation stage of the systemic ap-

proach of the scientiÞc thought. I show that it is the beginning of an uniÞ-
cation stage but remain to create synergies with the rest of the sciences.

Let {xt }t=∞

t=−∞ be a covariance-stationary process with mean E [xt ] = µ and
jth covariance γ j

E [(xt − µ)(xt−j − µ)] = γ j (Q16)

If these autocovariances are absolutely summable, the population spectrum
of x is given by [21] :
( )
1 X∞
Sx (w) = γ0 + 2 γ j cos(jw) (Q17)
2π j=1

27
 If the γ j ’s represents autocovariances of a covariance-stationary process and
using the Possibility Theorem, then

Sx (w) ∈ [Sx (w − σ w ), Sx(w + σ w )] (Q18)

and Sx (w) will be nonnegative for all w. In general for an ARMA(p,q) process

xt = c + φ1 xt−1 + φ2 xt−2 + ... + φp xt−p + εt +

θ1 εt−1 + θ2 εt−2 + ... + θq εt−q
½ 2 ¾
σ for t = τ
E [εt ετ ] = (Q19)
0 otherwise

The population spectrum Sx (w) ∈ [Sx (w − σ w ), Sx (w + σ w )] is given by [21].

σ 2 (1 + θ1 e−iw + θ2 e−i2w + ... + θq e−iqw )

Sx (w) =
2π (1 − φ1 e−iw − φ2 e−i2w − ... − φq e−iqw )
(1 + θ1 eiw + θ2 ei2w + ... + θq eiqw )
∗ (Q20)
(1 − φ1 eiw − φ2 ei2w − ... − φq eiqw )
√
where i = −1 and w is a scalar.
Given an observed sample of T observations denoted x1 , x2 , xT we can calcu-
late up to T − 1 sample autocovariances γ j from the formulas:

1 X
T
γbj = (xt − x) (xt−j − x) for j = 0, 1, ..T − 1
T t=j+1

1X
T
x = xt (Q21)
T t=1

The sample periodogram can be expressed as:

( )
1 X
T −1
Sbx (w) = b
γ0 + 2 b
γ j cos(jw) (Q22)
2π j=1

When the sample size T is an odd number, xt will be expressed in terms of

periodic functions with M = (T − 1)/2 different frequencies wj = ( 2π
T
)j < π.
M n
X o
b+
xt = µ baj cos(wj (t − 1)) + bbj sin(wj (t − 1)) + ut (Q23)
j=1

aj , bbj can be estimated with OLS regression.

The coefficients b

28
 2X
T
b
aj = xt cos(wj (t − 1)) j = 1, 2, ..., M
T t=1

2X
T
bbj = xt sin(wj (t − 1)) j = 1, 2, ..., M (Q24)
T t=1

The sample variance of xt can be expressed as:

1X
T
1 X 2 b2
M
2
(xt − x) = (b
a + bj ) (Q25)
T t=1 2 j=1 j

The portion of the sample variance of xt that can be attributed to cycles of

frequency wj is given by:
1 2 b2 4π b
aj + bj ) =
(b Sx(wj ) (Q26)
2 T
with Sbx (wj ) the sample periodogram of frequency wj .
Continuing with the methodology proposed by Hamilton [21]. we develop two
examples that will allow us to verify the applicability of the possibility theorem.

Example 2 As Þrst step we build four time series knowing all their parameters
aj , bj and wj = ( 2π
T
)j, j = 1, ..., M

(Table 7, Figures 2,3,4,5).

xt = 2 cos(w1 (t − 1)) − 3 cos(w3 (t − 1))

+2 sin(w5 (t − 1)) + 4 sin w1 (t − 1))
yt = 2 cos(w1 (t − 1)) + 4 sin(w12 (t − 1))
zt = 4 sin(w12 (t − 1))
ut = 3

Second step, we Þnd the value of the parameters γbj , j = 1, ..., (T − 1) for the
time series xt according to Q21 (Table 7, Figure 6).
Third step, we Þnd Sbx (wj ) (Table 8, Figure 7), only for the frequencies
w1 , w3 , w5 , w12 :
Fourth step, we compute the values σ x , σ w and σ w min = 2σ1x . resp. for y, z, u,
(Table 9).

29
 1X
T
σx = (xt − x)2
T t=1
1X
4
σw = (wj − w)2
4 j=1
1
Fifth step, we verify that possibility theorem is respected, σ x σ w ≥ 2
resp for
y, z, u, (Table 9).
Example 3 Let time series vt

Table 7: Time Series Values of xt , yt , zt , ut

j=t γj ρj wj γj*cos(w1*j) γj*cos(w3*j) γj*cos(w5*j) γj*cos(w12*j) xt yt zt ut cos(w1(t-1)) cos(w3(t-1)) sen(w5(t-1)) sen(w12(t-1))

1 3,5566 0,2097 0,1698 3,5055 3,1050 2,3498 -1,6010 -1,0000 2,0000 0,0000 3 1,0000 1,0000 0,0000 0,0000
2 -0,0304 -0,0018 0,3396 -0,0287 -0,0160 0,0039 0,0181 4,4253 5,5430 3,5718 3 0,9856 0,8730 0,7507 0,8930
3 8,5707 0,5054 0,5094 7,4824 0,3641 -7,1006 8,4472 -0,9191 -1,3299 -3,2157 3 0,9429 0,5243 0,9919 -0,8039
4 -5,0762 -0,2993 0,6792 -3,9495 2,2851 4,9125 1,4853 2,0619 1,0693 -0,6767 3 0,8730 0,0425 0,5600 -0,1692
5 -9,1882 -0,5418 0,8491 -6,0706 7,6121 4,1376 6,6354 6,2277 5,3810 3,8249 3 0,7780 -0,4502 -0,2519 0,9562
6 4,1666 0,2457 1,0189 2,1847 -4,1515 1,5529 3,9281 -0,7458 -1,4455 -2,7669 3 0,6607 -0,8285 -0,8929 -0,6917
7 -4,0303 -0,2377 1,1887 -1,5028 3,6727 -3,7998 0,5102 0,8480 -0,2853 -1,3339 3 0,5243 -0,9964 -0,9279 -0,3335
8 -8,3009 -0,4895 1,3585 -1,7492 4,9368 -7,2476 6,8796 6,7808 4,7136 3,9678 3 0,3729 -0,9113 -0,3333 0,9920
9 5,6071 0,3307 1,5283 0,2382 -0,7130 1,1825 4,8936 0,9424 -1,8169 -2,2383 3 0,2107 -0,5947 0,4875 -0,5596
10 -1,3549 -0,0799 1,6981 0,1720 -0,5050 0,8054 -0,0583 0,4688 -1,8677 -1,9527 3 0,0425 -0,1272 0,9775 -0,4882
11 -7,6012 -0,4482 1,8679 2,2254 -5,9132 7,5737 6,9285 4,2325 3,7424 3,9963 3 -0,1270 0,3727 0,8041 0,9991
12 7,3787 0,4351 2,0377 -3,3215 7,2723 -5,3286 5,7376 -4,3944 -2,2307 -1,6452 3 -0,2928 0,7779 0,0851 -0,4113
13 3,5415 0,2088 2,2075 -2,1057 3,3394 0,1492 0,7488 -7,7556 -3,4154 -2,5151 3 -0,4502 0,9856 -0,6917 -0,6288
14 -4,7306 -0,2790 2,3774 3,4151 -3,1261 -3,6795 4,5790 -2,1067 2,7204 3,9096 3 -0,5946 0,9429 -0,9991 0,9774
15 7,3117 0,4312 2,5472 -6,0575 1,5420 7,2070 4,8257 -5,6880 -2,4485 -1,0047 3 -0,7219 0,6608 -0,6285 -0,2512
16 0,9479 0,0559 2,7170 -0,8637 -0,2773 0,4973 0,3543 -4,9574 -4,6620 -3,0051 3 -0,8285 0,2109 0,1686 -0,7513
17 -12,9805 -0,7655 2,8868 12,5614 9,3691 3,7957 12,9349 4,4681 1,8877 3,7101 3 -0,9112 -0,2926 0,8513 0,9275
18 -3,0598 -0,1804 3,0566 3,0488 2,9609 2,7876 -1,6015 1,8073 -2,2706 -0,3352 3 -0,9677 -0,7218 0,9563 -0,0838
19 -4,8565 -0,2864 - 4,8390 4,7001 4,4263 -2,5510 -1,6735 -5,4012 -3,4084 3 -0,9964 -0,9677 0,4123 -0,8521
20 -14,5155 -0,8560 - 14,0475 10,4826 4,2574 14,4617 3,4914 1,4109 3,4037 3 -0,9964 -0,9678 -0,4115 0,8509
21 0,7591 0,0448 - -0,6918 -0,2225 0,3977 0,2822 -1,3371 -1,5915 0,3440 3 -0,9678 -0,7222 -0,9560 0,0860
22 5,3938 0,3181 - -4,4692 1,1346 5,3158 3,5689 -6,3602 -5,5360 -3,7134 3 -0,9113 -0,2931 -0,8518 -0,9284
23 -4,5998 -0,2712 - 3,3212 -3,0378 -3,5805 4,4498 0,3719 1,3420 2,9992 3 -0,8286 0,2104 -0,1695 0,7498
24 6,9863 0,4120 - -4,1550 6,5864 0,3007 1,4620 -1,1565 -0,4308 1,0133 3 -0,7220 0,6604 0,6278 0,2533
25 8,7250 0,5145 - -3,9290 8,6000 -6,2953 6,7966 -5,9310 -5,1009 -3,9114 3 -0,5947 0,9428 0,9991 -0,9779
26 -5,6106 -0,3309 - 1,6436 -4,3666 5,5907 5,1089 0,0353 1,6076 2,5082 3 -0,4503 0,9857 0,6924 0,6271
27 3,3472 0,1974 - -0,4256 1,2493 -1,9922 0,1366 -1,4358 1,0674 1,6533 3 -0,2929 0,7783 -0,0842 0,4133
28 6,0537 0,3570 - 0,2561 -0,7664 1,2712 5,2899 -6,9779 -4,2510 -3,9967 3 -0,1272 0,3732 -0,8036 -0,9992
29 -8,4908 -0,5007 - -1,7876 5,0459 -7,4096 7,0264 0,4539 2,0295 1,9449 3 0,0423 -0,1266 -0,9777 0,4862
30 0,7532 0,0444 - 0,2807 -0,6862 0,7103 -0,0970 3,4729 2,6667 2,2457 3 0,2105 -0,5943 -0,4883 0,5614
31 4,3535 0,2567 - 2,2820 -4,3380 1,6264 4,1075 0,1767 -3,2212 -3,9667 3 0,3727 -0,9110 0,3324 -0,9917
32 -10,8746 -0,6413 - -7,1833 9,0127 4,8880 7,8365 7,2184 2,3739 1,3255 3 0,5242 -0,9964 0,9276 0,3314
33 -0,9354 -0,0552 - -0,7277 0,4215 0,9050 0,2757 8,3673 4,0944 2,7733 3 0,6606 -0,8288 0,8933 0,6933
34 4,4848 0,2645 - 3,9149 0,1881 -3,7179 4,4218 -0,4090 -2,2665 -3,8223 3 0,7779 -0,4506 0,2528 -0,9556
35 -9,8590 -0,5814 - -9,2952 -5,1647 1,2608 5,8458 1,1695 2,4138 0,6679 3 0,8729 0,0419 -0,5593 0,1670
36 0,0000 - - - - - - 1,5515 5,1066 3,2210 3 0,9428 0,5239 -0,9918 0,8052
37 - - - - - - - -5,7174 -1,5966 -3,5678 3 0,9856 0,8728 -0,7513 -0,8919

Table 8: Frequencies, variances and sample periodogram of xt

Analysis of xt

Variance: sx2
E[(xt-E[xt])2] 16,9578
2 2 2 2
(a1 +a3 +b5 +b12 )/2 16,5000
γ0 16,9578
T 37

Frequencies Wavelength Regression Coefficients (aj2+bj2)/2 Sample Periodogram (4*pj/T)sx(wj)

2
w1 0,1698 l1 37,0000 a1 2,0000 a1 2,0000 Sx(w1) 4,9607 1,68474894
2
w3 0,5094 l2 12,3333 a3 -3,0000 a3 4,5000 Sx(w3) 21,9880 7,46761264
2
w5 0,8491 l3 7,4000 b5 2,0000 b5 2,0000 Sx(w5) 8,3504 2,83598419
2
w12 2,0377 l4 3,0833 b12 4,0000 b12 8,0000 Sx(w12) 45,3754 15,4104557

30
31
Table 9: Verification of Possibily Theorem for series xt , yt , zt , ut
Heisenberg's Uncertainty Principle

sxsw 3,3475 sysw 4,2344

2 2
E[x t] 0,0002 sx=E[(x t-E[xt]) ] 4,1180 E[yt] 0,0001 sy=E[(yt-E[yt]) ] 3,2059
2 2
E[w] 0,8915 sw=E[(w-E[w]) ] 0,8129 E[w] 1,1038 sw=E[(w-E[w]) ] 1,3208

swmin 0,1214 swmin 0,1560

Lower[wj] Min[j] Upper[wj] Max[j] Lower[wj] Min[j] Upper[wj] Max[j]

E[w1] 0,1698 0,0484 0,2850 0,2912 1,7150
E[w3] 0,5094 0,3880 2,2850 0,6309 3,7150
E[w5] 0,8491 0,7276 4,2850 0,9705 5,7150 E[w1] 0,1698 0,0138 0,0816 0,3258 1,9184
E[w12] 2,0377 1,9163 11,2850 2,1591 12,7150 E[w12 ] 2,0377 1,8818 11,0816 2,1937 12,9184

szsw 0,0000 susw 0,0000

2 2
E[zt] 0,0001 sz=E[(zt-E[zt]) ] 2,8674 E[ut] 3,0000 su=E[(ut-E[ut]) ] 0,0000
E[w12] 2,0377 sw=E[(w-E[w]) ]
2
− E[w] ∞ sw=E[(w-E[w]) ]
2
−

swmin 0,1744 swmin ∞

Lower[wj] Min[j] Upper[wj] Max[j] Lower[wj] Min[j] Upper[wj] Max[j]

E[w12] 2,0377 1,8634 10,9731 2,0377 12,0000 E[wi] 2,0377 0 0∞ ∞

yt Vs t
8,0000
6,0000
4,0000
2,0000
dd
yt

0,0000
-2,0000 0 5 10 15 20 25 30 35 40
-4,0000
-6,0000
-8,0000 Time

Figure 2 yt = 2 cos(w1 (t − 1)) + 4 sin(w12 (t − 1))

zt Vs t
5,0000
4,0000
3,0000
2,0000
1,0000
0,0000 dd
zt

-1,0000 0 5 10 15 20 25 30 35 40
-2,0000
-3,0000
-4,0000
-5,0000 Time

Figure 3 zt = 4 sin(w12 (t − 1))

32
 ut Vs t
3,5
3
2,5
2
dd
ut

1,5
1
0,5
0
0 5 10 15 20
Time 25 30 35 40

Figure 4 ut = 3

10,0000
Time Serie xt

8,0000

6,0000

4,0000

2,0000

0,0000
xt

0 5 10 15 20 25 30 35 40
-2,0000

-4,0000

-6,0000

-8,0000

-10,0000
t

xt cos(w1(t-1)) cos(w3(t-1)) sen(w5(t-1)) sen(w12(t-1))

Figure 5 Constitutive elements of time series xt

0,6000 Autocorrelations of xt Vs Lag(j)

0,4000

0,2000
Autocorrelations

0,0000

-0,2000 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37

-0,4000

-0,6000

-0,8000

-1,0000
Lag(j)

Figure 6 Autocorrelations of time series xt

33
 Figure 7 Peridiogram of time series xt

5 Conclusion
1 Hermite’s polynomials allow us to study the probability density function
of the vNM utility inside of a n− player game. Using the approach of Quantum
Mechanics we have obtained an equivalence between quantum level and strategy.
The function of states of Quantum Mechanics opens a new focus in the theory of
games such as sub-strategies.
2 An immediate application of quantum games in economics, is in direc-
tion to principal-agent relationship. SpeciÞcally we can use m types of agents for
the case of adverse selection. In moral risk quantum games could be used for a
discrete or continuous set of efforts.
3 In this paper we have demonstrated that the Nash-Hayeck equilibrium
opens new doors to the use of entropy in game theory. Without forgetting that
the main way to evidence the Nash’s equilibria is utility maximization, we can
affirm that the human behavior arbitrates between these two elements stochastic-
utility (beneÞts) U (p(x)) and entropy (risk or order) H(p(x)). Accepting that
U (p(x))
the relationship stochastic-utility/entropy H(p(x)) is an equivalent relationship
to the ³well-known´ beneÞts/cost we present a new way to calculate equilibria:
U (p(x))
M axx H(p(x)) , p(x) : probability function and x = (x1 , x2 , ..., xn ) exogenous
variables.
4 In all time series xt , where cycles are present, it is impossible to
know the exact value of the frequency wj of each one of the cycles j. We
can know the value of frequency in a certain conÞdence interval given by wj ∈
[wj − σ w min , wj + σ w min ] where σ w min = 2σ1x . “The more precisely the aleatory
variable VALUE x is determined, the less precisely the frequency VALUE w is
known in this instant, and conversely”

34
 5 Using Game Theory concepts we can obtain the equilibrium condition
in time series. The equilibrium condition in time series is represented by the
Possibility Theorem (Tables 7,8,9 and Figures 3,4,5,6,7).

ACKNOWLEDGMENT
This work was developped with Þnancial support of TODO1, www.todo1.com
and PETROECUADOR, www.petroecuador.com.ec.
The author is grateful for helpful comments to: Jorge Moyano from Diners
Club International, Gonzalo Pozo from Todo1 Services-Miami, the assistants of
8th Conference in Computing in Economics and Finance, July 2002, AIX France,
the assistants to 10th Annual Conference of the SouthEast SAS° R
User Group
USA, the assistants to M2002, the 5th annual data mining technology conference
SAS° R
2002 USA. The opinions and errors are solely my responsability. The au-
thor also thanks to Prof. Jean Louis Rulliere, GATE director, Universite Lumiere
Lyon II France, where he was Þnishing his PhD dissertation.
Corresponding Author. Fax 59322524766, email: ejimenezecu@yahoo.com

References
[1] Allais M. An Outline of My Main Contributions To Economic Science. AER,
Vol 87 No 5: 3-9, 1988.
[2] Anderson S.P., Goere J.K. Minimum effort Coordination Games: Stochastic
Potential and Logit Equilibrium. Games and Economics Behavior. 34:177-
199, 2001.
[3] Arrow K. Mathematical Models in The Social Sciences. General Systems,
29-47, 1956.
[4] Bacry H. Introduction aux concepts de la Mecanique Statistique. Ellipsis,
París,1991.
[5] Bertalanffy L. Teoría General de los Sistemas. Fondo de Cultura Económico.
USA, 1976.
[6] Binmore K. Fun and Games. D.C.Health and Company,1992
[7] Cohen-Tannoudji C., Diu B., Laloe F. Quantum Mechanics. Hermann,
París,1977.
[8] Eisert J., Scheel S., Plenio M.B. On the Impossibility of distilling Gaussian
States with Gaussian Operations. Mimeo, Imperial College of London, 2002.
[9] Eisert J. and Wilkens M. Quantum Games. Mimeo, University of Potsdam
of Germany, 2002.

35
[10] Eisert J., Wilkens M., Lewenstein M. Quantum Games and Quantum Strate-
gies, Working Paper, University of Postdam, Germany, 2001.

[11] Feynman R. Le cours de Physique de Feynman. DUNOT, 1998.

[12] Fleming W., Rishel R. Deterministic and Stochastic Optimal Control.1975.

[13] Frieden B. Information Sciences, Probability, Statistical, Optics and Data

Testing. Springer-Verlag.

[14] Fudenberg D., Tirole J. Game Theory, Massachusetts, The Mit Press, 1991.

[15] Geanakoplos J. Common Knowledge,Handbook of Game Theory. Vol II,

Chap 40, 1994.

[16] Gihman I., Skorohod A. Controlled Stochastic Processes. Springer Verlag,

New York, 1979.

[17] Gourieux Ch. Econometrie des Variables Qualitatives. 2éme-ed, París,

Económica, 1989.

[18] Griffel, D.H. Applied Functional Analysis. John Wiley & Sons, New York,
1985.

[19] Griliches Z., Intriligator M. Handbook of Econometrics, Vol 1, Elsevier, 1996.

[20] Hahn F. Stability, Handbook of Mathematical Economics. Vol II, Chap 16,
1982.

[21] Hamilton J.D. Time Series Analysis, Princeton University Press, New Jer-
sey, 1994, Chap(6).

[22] Hammerstein P., Selten R. Game Theory and Evolutionary Biology., Hand-
book of Game Theory. Vol 2, Elsevier B.V., 1994.

[23] Harsanyi J. Game and Desicion Theoreticac Models in Ethics,Handbook of

Game Theory. Vol I, Chap 19, 1992.

[24] Jiangfeng, Du and all . Experimental Realizations of Quantum Games on a

Quantum Computer, Mimeo, University of Science and Technology of China,
2002.

[25] Jimenez E. Mémoire DEA microéconomie appliquée., Université Lumière

Lyon II, GATE, Lyon, 2001.

[26] Jiménez E. UniÞed Game Theory (unpublished), Paper presented in 8th

CONFERENCE IN COMPUTING IN ECONOMICS AND FINANCE AIX
2002, 2002.

36
[27] Judge G., Griffiths W.E., Hill, Lütkepohl H., Lee T. The Theory and Practice
of Econometrics, John Wiley and sons., 1985.

[28] Kay, Lai Chung. (1978). A Course in Probability Theory, Harcourt, Brace &
World, INC.

[29] Kamien M., Schwartz N. Dynamic Optimization., Elsevier, Amsterdam,

1998.

[30] Karatzas I., Shreve S. Brownian Motion and Stochastic Calculus., Second
Edition, Springer-Verlag, 1991.

[31] Karlin S., Taylor H. A First Course in Stochastic Processes., Academic Press
Inc, Usa, 1980.

[32] Karlin S., Taylor H. A Second Course in Stochastic Processes., Academic

Press Inc, Usa, 1981.

[33] Kloeden P., Platen E. Numerical Solution of Stochastic Differential Equa-

tions, Springer-Verlag, 1995.

[34] Krishna V. and Sjostrom T. On he Convergence of Fictitious Play., Working

Paper, Harvard University, 1995.

[35] Landau. L (1978). Mecanica Cuántica, Editorial Reverté, 1978.

[36] Lévy E. Dictionnaire de Physique., PUF, 1998.

[37] Malliaris A., Brock W. Stochastic Methods in Economics in Finance., Else-

vier, Netherlands, 1982.

[38] McFadden D. Quantal Choice Analysis : A survey , Annals of Economicals

and Social Measurement 5/4., p.363-389, 1976.

[39] McFadden D. Econometric Analysis of Qualitative Response Models., Hand-

book of Econometrics Vol II, Chap 24, 1984.

[40] McFadden D. Economic Choices., AER, Vol 91 NO.3, 351-378, 2001.

[41] McKelvey R., Palfrey T. Quantal Response Equilibria for Normal Form
Games, Games and economics Behavior., 10:6-38, 1995.

[42] McKelvey R., Palfrey T. Quantal Response Equilibria., California Institute

of Technology, current version 1996.

[43] McKelvey R., Palfrey, T.), Quantal Response Equilibria for Extensive Form
Games. Experimental Economics, 1,9-41, 1998.

37
[44] Meyer D. Quantum Strategies. Phys. Rev. Lett. 82 (1999) 1052-1055, 1999.

[45] Meyer D. Quantum Games and Quantum Algorithms, To appear in the AMS
Contemporany Mathematics volume: Quantum Computation and Quantum
Information Science, 2000.

[46] Monderer D. and Sharpley LL. Fictitious Play Property for Games with
Identical Interests., Journal of Economic Theory 68: 258-265, 1996.

[47] Myerson R. Game Theory Analysis of Conßict., Massachusetts, Harvard

University Press, 1991.

[48] Neyman A. Repeated Games with Bounded Entropy., Games and Economic
Behavior, Ideal, 30, 228-247, 2000.

[49] NGÖ C., NGÖ H. Introduction à la Mecanique Statistique., Masson, 2éme

édition, 1995.

[50] Ochs J. Games with unique mixed strategy equilibria: An experimental

study,mimeo., University of Pittsburgh, 1993.

[51] Pena, L (1979). Introducción a la Mecánica Cuántica, Compania Editorial

Continental, México, 1979.

[52] Rasmusen E. Games and Economic Information: An Introduction to Game

Theory., Oxfort: Basil Blackwell, 1989.

[53] Rényl A. Calcul des Probabilités., Paris, Dunod, 1966.

[54] Robert Ch. ,L’Analyse statistique Bayesienne., Economica, 1992.

[55] Rubinstein A. The Electronic Mail Game: Strategic Behavior Under, Almost
Common Knowledge., American Economic Review, 79:385-391, 1989.

[56] Rullière J. L., Walliser B. De la spécularité à la temporalité en thèorie des

jeux., Revue d’Economie Politique, 105, pp 601-615, 1995.

[57] Rust J. Structural Estimation of Markov Decision Processes., Handbook of

Econometrics, Volume IV, Chapter 51, 1994.

[58] SENKO Corporation. Quantum Computer Simulation, SENKO Corporation,

Yokohama Japan, 1999.

[59] Simon H. A. Reason in human affairs., Oxfort blackwell., 1983.

[60] Suijs J., Borm.P. Stochastic Cooperative Games: Superadditivity Convexity

and Certainty Equivalents, Games and Economic Behavior., Ideal, 27, 331-
345, 1999.

38
[61] Tirole J. Théorie de L’organisation Industrielle., Tome II, Economica, 1988.

[62] Varian H. Analyse Microéconomique., Troisième édition, De Boeck Univer-

sité, Bruxelles, 1995.

[63] Weibul J.W. Evolutionary Game Theory., The MIT press, 1995.

[64] Lars-Goran Mattsson, Weibull J.W. Effort-based probabilistic choice., 2001.

[65] Wilson R. Strategic Analysis of Auctions., Handbook of Game Theory, Vol

I, Chap 8, 1992a.

[66] Wilson R. Strategic Models of entry deterrence., Handbook of Game Theory,

Vol I, Chap 10, 1992b

[67] Zauner K. A Payofff Uncertainty Explanation of Results in Experimental

Centipede Games., Games and Economic Behavior, 26, 157-185, 1999.

39