Analysis of Discrete Stochastic Population Models with Normal Distribution

Haiyan Wang · 2025

Read Analysis of Discrete Stochastic Population Models with Normal Distribution by Haiyan Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-14296.txt.

Introduction

Haiyan Wang analyzes a random population model in a 2025 mathematics paper. The model is a logistic difference equation, where growth depends on current size. Wang assumes the population follows a normal distribution.

The paper links average growth to a new concept called the uniform structural growth rate. That rate captures how the inner spread of the population shapes growth. Wang derives formulas tying it to the spread and to small random shocks. Two branches of the rate can exist, tied to higher and lower growth states. Gaps where no such rate exists are also proved.

Catalog record

CREATORHaiyan Wang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-14296

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Analysis of Discrete Stochastic Population Models with Normal Distribution Haiyan Wang Abstract This paper analyzes a stochastic logistic difference equation under the assumption that the population distribution follows a normal distribution. Our focus is on the mathematical relationship between the average growth rate and a newly introduced concept—the uniform structural growth rate, which captures how growth is influenced by the internal distributional structure of the population. We derive explicit relationships linking the uniform structural growth rate to the parameters of the normal distribution and the variance of a small stochastic perturbation. The analysis reveals the existence of two distinct branches of the uniform structural growth rate, corresponding to alternative population states characterized by higher and lower growth rates. This duality provides deeper insights into the dynamics of population growth under stochastic influences. A sufficient condition for the existence of two uniform structural growth rates is established and rigorously proved, demonstrating that there exist infeasible intervals where no uniform structural growth rate can be defined. We also explore the biological significance of these findings, emphasizing the role of stochastic perturbations and the distribution in shaping population dynamics. 1 Introduction Logistic difference equations are commonly used to model population growth in environments with limited resources. In their deterministic form, these models capture an initial phase of exponential growth that gradually slows as the population nears the carrying capacity—the maximum population size that the environment can support. The behavior of these models is highly sensitive to the intrinsic growth rate, which governs the rate of expansion and the dynamics of the system [ 1 , 6 , 3 ] . Stochastic models provide important advantages over their deterministic counterparts. While deterministic models yield fixed outcomes based solely on initial conditions, stochastic models capture the full distribution of possible population sizes, offering deeper insights into both the expected dynamics and the variability around them. By incorporating randomness, these models more accurately reflect the behavior of real-world populations, especially under fluctuating environmental conditions. The study of stochastic equations has become a vibrant area of research, with extensive literature exploring a wide range of extensions and applications [ 2 , 5 , 3 , 8 , 4 , 7 , 11 , 10 ] . In this paper, we study the stochastic logistic equation from time t to t+1 X_{t+1}=rX_{t}\left(1-X_{t}\right)\epsilon_{t}, (1) where X_{t} is a distribution of population size at time t , r>0 is a real number, \epsilon_{t} is a small nonnegative perturbation distribution representing stochastic effects. \epsilon_{t} is independent of X_{t} and its mean is 1 ( E[\epsilon_{t}]=1 ). We assume that X_{t} follows a normal distribution ( X_{t}\sim N(\mu,\sigma^{2}),\mu>0,\sigma>0) . Here we call r a uniform intrinsic growth rate under structural transformation , abbreviated as the uniform structural growth rate — a constant applied uniformly across the entire of the domain of X_{t} , following a nonlinear structural transformation given by X_{t}(1-X_{t}) . This differs from an average growth rate defined by the expectation E[rX_{t}(1-X_{t})] , which captures overall outcomes but not underlying structure. We try to address this question: In order to achieve an average growth rate \alpha>0 (E[X_{t+1}]=\alpha E[X_{t}] ), what are the required uniform structural growth rate r ? Our analysis in this paper indicates that there are multiple branches of uniform structural growth rates, which represent alternative states corresponding to higher and lower growth rates, respectively. We establish the mathematical relationship between the uniform structural growth rate r and the parameters of the normal distribution. A sufficient condition is presented to guarantee the existence of two uniform structural growth rates. We demonstrate that there are unfeasible intervals of E[X_{t}] with which the population does not have a uniform structural growth rate to achieve (E[X_{t+1}]=\alpha E[X_{t}] . This duality provides deeper insights into population growth under stochastic conditions. We further investigate the impact of the parameters of the normal distribution and describe biological interpretations. The influence of the parameters of the normal distribution and perturbation \epsilon_{t} on r provides theoretical insights into how the expectation and variance of the underlying population impact uniform structural growth rates. The work extends the analysis in [ 9 ] where the gamma distribution is used to address the growth rate of the stochastic logistic and Ricker logistic equations at equilibrium. In this paper, the population is assumed to be in a state following a normal distribution. The study offers deeper insights into the interplay among stochastic effects, uniform structural growth rates, nonlinear transformation and distributional parameters, providing a more comprehensive framework for understanding real-world population dynamics. 2 Mathematical formulation Let’s state our major result on the relation of r with the parameters of the normal distribution as a theorem. Theorem 1 . For stochastic logistic equation ( 1 ), assume that X_{t} follows a normal distribution N(\mu,\sigma^{2}) , and E[\epsilon_{t}]=1,E[\epsilon_{t}^{2}]=v with parameters v>0 , \sigma^{2}>0 , 0<\mu<1 . In addition, we assume that \displaystyle E[X_{t+1}] \displaystyle=\alpha E[X_{t}], (2) \displaystyle\operatorname{Var}[X_{t+1}] \displaystyle=\beta\operatorname{Var}[X_{t}], (3) where \beta>0 , \alpha>0 . Then r in ( 1 ) are positive solutions of the following polynomial in terms of r : \displaystyle\begin{aligned} &v(-2\mu^{4}+4\mu^{3}-3\mu^{2}+\mu)r^{3}-v\alpha\mu r^{2}\\ &\quad+\left[v(3\alpha^{2}\mu^{2})+\beta\mu(\mu-1)-\alpha^{2}\mu^{2}\right]r+\beta\alpha\mu=0\end{aligned} (4) Proof. Moments of X : We have the following relations from Appendix 6 \displaystyle\begin{aligned} E[X_{t}]&=\mu,&E[X_{t}^{2}]&=\mu^{2}+\sigma^{2},\\ E[X_{t}^{3}]&=\mu^{3}+3\mu\sigma^{2},&E[X_{t}^{4}]&=\mu^{4}+6\mu^{2}\sigma^{2}+3\sigma^{4},\end{aligned} (5) Expectation condition : We begin with the expectation condition. Because of E[\epsilon_{t}]=1, we have \displaystyle\begin{aligned} E[X_{t+1}]&=rE[X_{t}(1-X_{t})],\\ &=r\left(\mu-(\mu^{2}+\sigma^{2})\right),\\ \end{aligned} (6) With ( 3 ), it follows that \displaystyle\begin{aligned} &r\left(\mu-\mu^{2}-\sigma^{2}\right)=\alpha\mu,\\ &\sigma^{2}=\mu-\mu^{2}-\frac{\alpha\mu}{r}.\end{aligned} (7) Variance condition : We continue with variance condition in ( 3 ) \displaystyle\operatorname{Var}[X_{t+1}] \displaystyle=E[X_{t+1}^{2}]-(E[X_{t+1}])^{2}=\beta\sigma^{2} (8) and \displaystyle E[X_{t+1}^{2}] \displaystyle=\beta\sigma^{2}+(\alpha\mu)^{2}, (9) On the other hand, from ( 1 ), we have \displaystyle\begin{aligned} E[X_{t+1}^{2}]&=r^{2}E[\epsilon_{t}^{2}]E\left[X_{t}^{2}(1-X_{t})^{2}\right],\\ &=r^{2}vE\left[X_{t}^{2}-2X_{t}^{3}+X_{t}^{4}\right]\end{aligned} (10) In view of ( 5 ) \displaystyle\begin{aligned} E[X_{t+1}^{2}]&=r^{2}v\left[(\mu^{2}+\sigma^{2})-2(\mu^{3}+3\mu\sigma^{2})\right.\\ &\quad\left.+(\mu^{4}+6\mu^{2}\sigma^{2}+3\sigma^{4})\right],\\ &=r^{2}v\left[\mu^{4}-2\mu^{3}+\mu^{2}+(6\mu^{2}\sigma^{2}-6\mu\sigma^{2}+\sigma^{2})+3\sigma^{4}\right].\end{aligned} (11) Thus we arrive at this relation with all the parameters. \displaystyle r^{2}v\left[\mu^{4}-2\mu^{3}+\mu^{2}+(6\mu^{2}\sigma^{2}-6\mu\sigma^{2}+\sigma^{2})+3\sigma^{4}\right] \displaystyle=\beta\sigma^{2}+(\alpha\mu)^{2}, (12) We will eliminate \sigma using ( 7 ). Note that \displaystyle\begin{aligned} \sigma^{4}&=\left(\mu-\mu^{2}-\frac{\alpha\mu}{r}\right)^{2},\\ &=\left(\mu-\mu^{2}\right)^{2}-2\left(\mu-\mu^{2}\right)\frac{\alpha\mu}{r}+\frac{\alpha^{2}\mu^{2}}{r^{2}},\\ &=\mu^{2}-2\mu^{3}+\mu^{4}-\frac{2\alpha\mu^{2}}{r}+\frac{2\alpha\mu^{3}}{r}+\frac{\alpha^{2}\mu^{2}}{r^{2}},\\ \end{aligned} (13) Now substitute \sigma^{2} and \sigma^{4} into the left side of ( 12 ): \displaystyle\begin{aligned} &vr^{2}\left[\mu^{4}-2\mu^{3}+\mu^{2}+(6\mu^{2}-6\mu+1)\sigma^{2}+3\sigma^{4}\right],\\ &=vr^{2}\left[\mu^{4}-2\mu^{3}+\mu^{2}\right.\\ &\quad+(6\mu^{2}-6\mu+1)\left(\mu-\mu^{2}-\frac{\alpha\mu}{r}\right)\\ &\quad\left.+3\left(\mu^{4}-2\mu^{3}+\mu^{2}-\frac{2\alpha\mu^{2}}{r}+\frac{2\alpha\mu^{3}}{r}+\frac{\alpha^{2}\mu^{2}}{r^{2}}\right)\right],\\ \end{aligned} (14) Expanding the terms in the left side of ( 14 ), the left side of ( 12 ) becomes \displaystyle\begin{aligned} &\frac{1}{r}vr^{3}\left[\mu^{4}-2\mu^{3}+\mu^{2}\right.\\ &\quad+\left(6\mu^{3}-6\mu^{4}-\frac{6\alpha\mu^{3}}{r}-6\mu^{2}+6\mu^{3}+\frac{6\alpha\mu^{2}}{r}+\mu-\mu^{2}-\frac{\alpha\mu}{r}\right)\\ &\quad+\left(3\mu^{4}-6\mu^{3}+3\mu^{2}-\frac{6\alpha\mu^{2}}{r}+\frac{6\alpha\mu^{3}}{r}+\frac{3\alpha^{2}\mu^{2}}{r^{2}}\right)\bigg],\\ &\quad=\frac{1}{r}vr^{3}\left[-2\mu^{4}+4\mu^{3}-3\mu^{2}+\mu-\alpha\mu\frac{1}{r}+3\alpha^{2}\mu^{2}\frac{1}{r^{2}}\right],\\ &\quad=\frac{1}{r}v\left[-2\mu^{4}r^{3}+4\mu^{3}r^{3}-3\mu^{2}r^{3}+\mu r^{3}-\alpha\mu r^{2}+3\alpha^{2}\mu^{2}r\right].\end{aligned} (15) On the other hand, the right side of ( 12 ) is \displaystyle\begin{aligned} &\beta\sigma^{2}+(\alpha\mu)^{2},\\ &=\beta\left(\mu-\mu^{2}-\frac{\alpha\mu}{r}\right)+\alpha^{2}\mu^{2},\\ &=\beta\mu-\beta\mu^{2}-\frac{\beta\alpha\mu}{r}+\alpha^{2}\mu^{2},\\ &=\frac{1}{r}\left[\beta\mu r-\beta\mu^{2}r-\beta\alpha\mu+\alpha^{2}\mu^{2}r\right],\\ \end{aligned} (16) Combining ( 12 ), ( 15 ) and ( 16 ) we arrive at the following simplified equation without \sigma after cancelling \frac{1}{r} at the both sides. \displaystyle\begin{aligned} v&\left[-2\mu^{4}r^{3}+4\mu^{3}r^{3}-3\mu^{2}r^{3}+\mu r^{3}-\alpha\mu r^{2}+3\alpha^{2}\mu^{2}r\right]\\ &=\beta\mu r-\beta\mu^{2}r-\beta\alpha\mu+\alpha^{2}\mu^{2}r,\end{aligned} (17) Combining the like terms for r^{i},i=3,2,1,0 , we conclude the proof of Theorem. \displaystyle\begin{aligned} &v(-2\mu^{4}+4\mu^{3}-3\mu^{2}+\mu)r^{3}-v\alpha\mu r^{2}\\ &\quad+\left[v(3\alpha^{2}\mu^{2})+\beta\mu(\mu-1)-\alpha^{2}\mu^{2}\right]r+\beta\alpha\mu=0\end{aligned} (18) ∎ 3 Conditions for two positive solutions Consider the cubic polynomial from the stochastic logistic model: \displaystyle p(r)=ar^{3}+br^{2}+cr+d=0, (19) where: \displaystyle a \displaystyle=v(-2\mu^{4}+4\mu^{3}-3\mu^{2}+\mu), (20) \displaystyle b \displaystyle=-v\alpha\mu, (21) \displaystyle c \displaystyle=v(3\alpha^{2}\mu^{2})+\beta\mu(\mu-1)-\alpha^{2}\mu^{2}, (22) \displaystyle d \displaystyle=\beta\alpha\mu, (23) with parameters v>0 , \beta>0 , \alpha>0 , 0<\mu<1 , and r as the growth rate. We seek conditions ensuring exactly two positive real roots and one negative root. Theorem 2 . Assume that parameters v>0 , \beta>0 , \alpha>0 , 0<\mu<1 . Then ( 19 ) has exactly two positive solutions if \displaystyle\Delta>0\quad\text{and}\quad p\left(\frac{2v\alpha\mu+\sqrt{\Delta}}{6a}\right)<0. (24) where \Delta=4v^{2}\alpha^{2}\mu^{2}-12ac . Proof. Claim a>0 : Let g(\mu)=(-2\mu^{3}+4\mu^{2}-3\mu+1) and \displaystyle a=v\mu(-2\mu^{3}+4\mu^{2}-3\mu+1)=v\mu g(\mu). (25) Note that g(0)=1,g(1)=0 and g^{\prime}(\mu)=-6\mu^{2}+8\mu-3 . The

Read the complete text