In this paper, we analyze a modified Susceptible‐Exposed‐Infectious‐Dead‐Recovered (SEIDR) model in the literature of the Ebola disease with uncertainties. The model is constructed using a van Kampen expansion method to have an Ebola SEIDR stochastic Fokker–Planck equation model. This model has a deterministic equation and noise covariance matrix. The basic reproduction number of the deterministic equation is calculated using the next‐generation matrix method. We prove the uniqueness and existence of the deterministic model using Lipschitz conditions and also show that it is locally asymptotically stable at it endemic equilibrium states. We constructed two equivalent stochastic differential equations (SDEs) models, whereas the Weiner process is equal to (i) the number of model equations and (ii) the number of independent changes in the model. Our aim is to solve (i) computationally using Cholesky decomposition technique with the variance–covariance matrix. Our proposed Cholesky decomposition SEIRD‐SDEs model is compared with (ii) and also with the epidemic data of the 2018 Ebola outbreak in the Democratic Republic of Congo. We also use numerical analyses to show that the importance of post‐death transmission is difficult to identify with noise terms supported by statistical hypothesis testing.

中文翻译：

---

使用Cholesky分解为2018年埃博拉病毒病暴发建模

在本文中，我们在不确定性的埃博拉病文献中分析了一种改良的易暴露，传染性，死亡恢复（SEIDR）模型。该模型是使用van Kampen展开方法构建的，具有Ebola SEIDR随机Fokker-Planck方程模型。该模型具有确定性方程式和噪声协方差矩阵。确定性方程式的基本复制数是使用下一代矩阵方法计算的。我们使用Lipschitz条件证明了确定性模型的唯一性和存在性，并且还表明了该模型在其地方均衡状态下是局部渐近稳定的。我们构造了两个等效的随机微分方程（SDE）模型，而Weiner过程等于（i）模型方程的数量和（ii）模型中独立变化的数量。我们的目标是通过方差-协方差矩阵使用Cholesky分解技术来求解（i）。我们将拟议的Cholesky分解SEIRD-SDEs模型与（ii）进行了比较，还与2018年刚果民主共和国埃博拉疫情的流行数据进行了比较。我们还使用数值分析表明，用统计假设检验支持的噪声项很难确定死亡后传播的重要性。