Dynamics of Quadratic Stochastic Operators with Some Coordinates
Abstract
In this article, we determine all fixed points of quadratic stochastic operators with invariant coordinates, and we conduct a comprehensive analysis of the local and global dynamics of the given operator. In particular, we investigate the stability of the fixed points and classify their types using appropriate dynamical system techniques. Furthermore, we study the asymptotic behavior of trajectories generated by the operator and establish conditions under which convergence occurs. Several illustrative examples are provided to demonstrate the theoretical results and to highlight different dynamical regimes, including stability, periodicity, and convergence to boundary points.
Keywords:Quadratic stochastic operatorFixed pointSimplexTrajectory
1. Introduction
One of the most actively developing directions of modern mathematics is the theory of dynamical systems. This theory studies the evolution of processes over time and is investigated in both discrete and continuous settings. Dynamical systems are widely applied in many fields such as biology, economics, physics, and information technology. Scientists such as Poincare, Lyapunov, and Kolmogorov played an important role in the formation of this theory.
The Uzbek mathematical school has also achieved significant results in this direction. In particular, stochastic operators, dynamical processes on simplices, and their stability properties have been deeply studied by T.A. Sarimsoqov, R.A. Ganikhodzhaev, N.N. Ganikhodzhaev, U.A. Rozikov, and U. Jamilov.
Let ( ) Sn−1 = x ∈ Rn : xi ≥ 0, i = 1, n, xi = 1
be the (n − 1)-dimensional simplex, and define an operator V : Sn−1 → Sn−1 by x′ k = Pij,kxixj, k = 1, . . . , n. (1)
i,j=1
Here,
Pij,k = Pji,k ≥ 0, Pij,k = 1. (2)
Definition 1. The operator defined by 1-2 is called a quadratic stochastic operator (QSO).
The condition Pij,k = 0, k / ∈ {i, j} (3) is called the Volterra condition, while Pij,k = 0, k ∈ {i, j} (4) is called the non-Volterra condition. If 3 holds, then the operator is called a Volterra QSO; if 4 holds, it is called a non-Volterra QSO.
Definition 2. Let V : Sn−1 → Sn−1 be a quadratic stochastic operator. A point x∗ ∈ Sn−1 is called a fixed point of V if V (x∗) = x∗.
We define a quadratic stochastic operator V : S2 → S2 as follows:
x′ k = Pij,kxixj, k = 1, 2,
i,j=1
x′ 3 = x3 = x3xj.
For this operator, we have x2 = 1 − x1 − x3, x1 = x, which yields the quadratic function f(x) = Ax2 + Bx + C. (6)
From the equality x′ 3 = x1x3 + x2x3 + x2 3, we obtain P33,3 = 1, P33,1 = P33,2 = 0, P23,3 = , P13,3 = , and P12,1 + P12,2 + P12,3 = 1.
Let us introduce the following notations:
x3 = R, P11,1 = a1, P12,1 = a2, P13,1 = a3, P22,1 = b1, P23,1 = b2, P33,1 = 0.
Then the quadratic function takes the form f(x) = (a1 − 2a2 + b1)x2 + 2 ((a2 − b1)(1 − R) + R(a3 − b2)) x + b1(1 − R)2 + 2b2R(1 − R).
Thus, A = a1 − 2a2 + b1, B = 2 ((a2 − b1)(1 − R) + R(a3 − b2)) , C = b1(1 − R)2 + 2b2R(1 − R), where A ∈ [−2, 2] \ {0}, B ∈ [−3, 3], C ∈ [0, 1].
2. Fixed Points
We begin by determining the fixed points of the function f(x) = Ax2 + Bx + C.
Theorem 1. If ( C ≤ A, B = 1 − 2√AC, then the function f(x) has a unique fixed point
x = ∈ [0, 1].
Proof : Fixed points satisfy Ax2 + Bx + C = x, which is equivalent to Ax2 + (B − 1)x + C = 0.
A unique solution exists when the discriminant is zero:
∆ = (B − 1)2 − 4AC = 0.
Hence, B = 1 ± 2√AC.
The case B = 1 + 2√AC yields a negative solution, while B = 1 − 2√AC gives √AC x = ∈ [0, 1], which holds if and only if 0 ≤ C ≤ A. □
Theorem 2. If the coefficients satisfy −2 ≤ A < 0, 2A + B < 1, A + B + C ≥ 1, then f(x) has a unique fixed point x1 = .
Theorem 3. Let 0 < A ≤ 2. Then the function f(x) has the fixed point
if 2A + B < 1, ( A + B + C ≤ 1, or B > 1 − 2√AC.
B > 1 − 2√AC, Moreover, it has the fixed point 1 − B + (B − 1)2 − 4AC
if B > 1 − 2√AC,
A + B + C ≥ 1.
3. Type of Fixed Points
Theorem 4. (1) If ∆ = 0, then the fixed point is a hyperbolic attracting fixed point.
(2) The fixed point x1 is hyperbolic attracting when ∆ < 4, and hyperbolic repelling when ∆ > 4.
(3) The fixed point x2 is a hyperbolic repelling fixed point.
Proof : i) Consider the case ∆ = 0. In this case, we use topological conjugacy. Let f(x) = Ax2 + Bx + C, g(x) = µx(1 − x), h(x) = mx + n.
We use the conjugacy relation h ◦ f = g ◦ h.
Then h ◦ f = mAx2 + mBx + (mC + n), g ◦ h = −µm2x2 + µm(1 − 2n)x + µ(n − n2).
Comparing coefficients, we obtain the system −µm = A, µ(1 − 2n) = B, 2 µ(n − n ) = mC + n.
Solving this system yields µ = 1, m = −A, n = .
This implies that the fixed point is attracting.
ii) It is known that for the fixed point
x1 = ,
The derivative satisfies f ′(x1) = 2Ax1 + B = 1 − √∆.
Therefore, the fixed point x1 is hyperbolic attracting when ∆ < 4 and hyperbolic repelling when ∆ > 4.
iii) For the fixed point 1 − B + (B − 1)2 − 4AC x2 = , we have f ′(x2) = 2Ax2 + B = 1 + √∆.
Hence, this fixed point is a hyperbolic repeller. □
Acknowledgments
The work is supported by the applied Project "Dynamics and Applications of Cubic Stochastic Operators"No. AL-9224093956 of The Agency for Innovative Development under the Ministry of Higher Education, Science and Innovation of the Republic of Uzbekistan.
Author Contribution
All authors have read and agreed to the published version of the manuscript.
Funding
The authors declare that no external funding or support was received for the research presented in this paper, including administrative, technical, or in-kind contributions. Data Availability All data supporting the reported findings in this research paper are provided within the manuscript.
Conflicts of Interest
The authors declare that there is no conflict of interest concerning the reported research findings. Funders played no role in the study’s design, in the collection, analysis, or interpretation of the data, in the writing of the manuscript, or in the decision to publish the results.
References
- [1] Devaney, R. (2018). An introduction to chaotic dynamical systems. CRC Press. https://doi.org/10.1201/9780429280801
- [2] Diyorov, A. M., & Rozikov, U. A. (2022). Evolutionary behavior in a two-locus system. https://doi.org/10.48550/arXiv.2210.00247
- [3] Ganikhodjaev, N. N., Pah, C. H., & Rozikov, U. (2021). Dynamics of quadratic stochastic operators generated by China's five element philosophy. Journal of difference equations and applications, 27(8), 1173-1192. https://doi.org/10.1080/10236198.2021.1968385
- [4] Ganikhodzhaev, R., Mukhamedov, F., & Rozikov, U. (2011). Quadratic stochastic operators and processes: Results and open problems. Infinite dimensional analysis, quantumprobability and related topics, 14(02), 279-335. https://doi.org/10.1142/S0219025711004365
- [5] Galor, O. (2007). Discrete dynamical systems. Springer Science & Business Media. https://doi.org/10.1007/3-540-36776-4
- [6] Lyubich, Y. I., Akin, E., Karpov, A. E., & Vulis, D. (1992). Mathematical structures in population genetics. Berlin: Springer-Verlag. https://link.springer.com/book/9783642762130
- [7] Mukhamedov, F., & Ganikhodjaev, N. (2015). Quantum quadratic operators and processes. Berlin: Springer. https://doi.org/10.1007/978-3-319-22837-2
- [8] Müller, J., & Kuttler, C. (2015). Methods and models in mathematical biology. Lecture Notes on Mathematical Modelling in Life Sciences, Springer, Berlin. https://doi.org/10.1007/978-3-642-27251-6
- [9] Rozikov, U. A. (2020). Population dynamics: Algebraic and probabilistic approach. World Scientific Publishing. https://doi.org/10.1142/9789811211232_0001
- [10] Rozikov, U. A., & Solaeva, M. N. (2023). Behavior of trajectories of a quadratic operator. Lobachevskii journal of mathe-matics, 44(7), 2910-2915. https://doi.org/10.1134/S1995080223070387