文章基本信息

标题：Feasibility of Singular Spectrum Analysis in the Field of Forecasting Mortality Rate
本地全文：下载
作者：Rahim Mahmoudvand ; Fatemeh Alehosseini ; Mohammad Zokaei 等
期刊名称：Journal of Data Science
印刷版ISSN：1680-743X
电子版ISSN：1683-8602
出版年度：2013
卷号：11
期号：4
页码：851-866
出版社：Tingmao Publish Company
摘要：The Lee-Carter model and its extensions are the most popular methodsin the eld of forecasting mortality rate. But, in spite of introducing severaldierent methods in forecasting mortality rate so far, there is no generalmethod applicable to all situations.Singular Spectrum Analysis (SSA) is a relatively new, powerful and nonparametrictime series analysis that its capability in forecasting dierent timeseries has been proven in the various sciences. In this paper, we investigatethe feasibility of using the SSA to construct mortality forecasts. We use theHyndman-Ullah model, which is a new extension of Lee-Carter model, as abenchmark to evaluate the performance of the SSA for mortality forecastsin France data sets.
关键词：Hyndman-Ullah model; Lee-Carter model; Mortality; Singular;Spectrum Analysis.