ABSTRACT

In a Covid-19 susceptible-infected-recovered-dead model with time-varying rates of transmission, recovery, and death, the parameters are constant in small time intervals. A posteriori parameters result from the Euler-Maruyama approximation for stochastic differential equations and from Bayes’ theorem. Parameter estimates and 10-day predictions are performed based on Moroccan and Italian Covid-19 data. Mean absolute errors and mean square errors indicate that predictions are of good quality.

中文翻译：

---

摩洛哥和意大利 Covid-19 流行期间基本繁殖数的贝叶斯预测

摘要

在 Covid-19 易感-感染-康复-死亡模型中，传播、恢复和死亡的速度随时间变化，参数在很小的时间间隔内保持不变。后验参数来自随机微分方程的 Euler-Maruyama 近似和贝叶斯定理。参数估计和 10 天预测是根据摩洛哥和意大利 Covid-19 数据进行的。平均绝对误差和均方误差表明预测质量良好。