Uncertainty and variability in demography and population growth: A hierarchical approach
Uncertainty and variability in demography and population growth: A hierarchical approach
复制标题
DOI：
10.1890/0012-9658(2003)084
复制
发表时间：
2003-06-01
期刊：
影响因子：
4.8
通讯作者：
Clark, JS
中科院分区：
文献类型：
--
作者：
Clark, JS
对不确定性的估计是推断人口风险的基础。不确定性是根据拟合到数据的模型来估计的，所述模型通常包括确定性模型（例如，人口增长）和随机因素，其中应考虑到抽样误差和影响观测的任何可变性来源。从拟合模型（比如人口统计学）到新变量（比如人口增长）的预测需要传播这些随机元素。生态模型忽略了大多数形式的变异性，因为它们使统计模型变得复杂，并带来了计算挑战。与空间，时间和个体之间的变异性，是不适应人口模型可以使参数估计和增长率的预测不切实际的，我适应一个层次的方法来估计人口增长率和他们的不确定性的问题时，个人的变化和变异性不能被分配到特定的原因。与为每个个体分配不同参数值的过拟合模型相反，分层模型适应个体差异，但假设这些差异来自潜在的分布-它们属于“总体”。分层模型可以在经典（频率论）和贝叶斯框架（我演示了两者）中实现，并使用马尔可夫链蒙特卡罗模拟进行分析。结果表明，依赖于估计误差的标准传播但忽略个体间变异性的人口增长模型可能会以损害可信度的方式歪曲不确定性。
Estimates of uncertainty are the basis for inference of population risk. Uncertainty is estimated from models fitted to data that typically include a deterministic model (e.g., population growth) and stochastic elements, which should accommodate errors in sampling and any sources of variability that affect observations. Prediction from fitted models (of, say, demography) to new variables (say, population growth) requires propagation of these stochastic elements. Ecological models ignore most forms of variability, because they make statistical models complex, and they pose computational challenges. Variability associated with space, time, and among individuals that is not accommodated by demographic models can make parameter estimates and growth rate predictions unrealistic.I adapt a hierarchical approach to the problem of estimating population growth rates and their uncertainties when individuals vary and that variability cannot be assigned to specific causes. In contrast to an overfitted model that would assign a different parameter value to each individual, hierarchical models accommodate individual differences, but assume that those differences derive from an underlying distribution-they belong to a "population." The hierarchical model can be implemented in classical (frequentist) and Bayesian frameworks (I demonstrate both) and analyzed using Markov chain Monte Carlo simulation. Results show that population growth models that rely on standard propagation of estimation error but ignore variability among individuals can misrepresent uncertainties in ways that erode credibility.