Coherent Modeling of the Risk in Mortality Projections: A Semi-Parametric Approach∗
Coherent Modeling of the Risk in Mortality Projections: A Semi-Parametric Approach∗
复制标题
死亡率预测风险的连贯建模：半参数方法*
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Daniel Bauer
中科院分区：
文献类型：
--
作者：
Nan Zhu;Daniel Bauer
与传统的死亡率风险建模方法不同，本文考虑的是死亡率预测的不确定性，而不是实际死亡率的不确定性。具体而言，我们对基于年度死亡率数据的滚动窗口和固定预测方法生成的死亡率预测的时间序列进行分析。我们发现，一到两个因素能够捕获数据中的绝大多数变化，并且这些因素的形状在不同的预测方法和人群中非常相似。依赖于一个半参数表示，包括所有的自洽模型的过渡参数化的正态分布的随机向量，我们确定和估计合适的模型的动态演变的死亡率预测。实例应用表明，由此产生的误差估计是相当大的不同，从传统的方法。因此，由此产生的模型提出了一致的，简约的，和听话的工作马的情况下，死亡率预测的风险评估是重要的。
Unlike conventional approaches to modeling mortality risk, this paper considers uncertainties in mortality projections rather than uncertainties in realized mortality rates. Specifically, we conduct our analysis on a time series of mortality forecasts generated based on a rolling window of annual mortality data and a fixed forecasting methodology. We find that one to two factors are capable of capturing the great majority of all the variation in the data, and that the shape of these factors is very similar across different forecasting methodologies and populations. Relying on a semi-parametric representation that encompasses all self-consistent models with transitions parameterized by Normal-Distributed random vectors, we identify and estimate suitable models for the dynamic evolution of mortality forecasts. Example applications show that resulting error estimates are considerably different from conventional approaches. Hence, the resulting models present coherent, parsimonious, and tractable workhorses in situations where the appraisal of risks within mortality projections is important.
死亡率的波动性
DOI：
10.2202/2153-3792.1035
发表时间：
2008
期刊：
Asia-Pacific Journal of Risk and Insurance
影响因子：
--
作者：
Daniel Bauer;Matthias Börger;Jochen Ruß;Hans-Joachim Zwiesler
通讯作者：
Hans-Joachim Zwiesler