English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

This paper evaluates the out-of-sample performance of two stochastic models used to forecast age specific mortality rates: (1) the model proposed by Lee and Carter (1992); and (2) a set of univariate autoregressions linked together by a common residual covariance matrix (Denton, Feavor, and Spencer 2005). To this aim, death rates from 16 industrialized nations are used to compare observed ex-post mortality rates to the forecasts generated by the models. Several functions of the individual age-specific mortality rates are also entertained, including life expectancy at birth (e0), as well as alternative measures of the age-dependency ratio. The latter are constructed based on how the individual mortality rates enter a population projection, and thus, are meant to gauge the potential impact of mortality alone on public retirement programs. In general, both models are found to produce point forecasts for the individual mortality rates, life expectancy, and the dependency ratios that are fairly close to one another. Typically, the median projections of mortality moderately overpredict the actual death rates, particularly for the oldestage groups (ages 65-95 or older). Conversely, the large majority of the point forecasts of life expectancy at birth and the dependency ratios underestimate their observed values. The models also generate interval forecasts of e0 that are "too wide" as their empirical probability content often exceeds their nominal coverage. However, the Lee-Carter model tends to seriously underpredict the forecast uncertainty associated with both the death rates of the oldest ages and the age-dependency ratios, while the autoregressive approach overpredicts this uncertainty in most cases.

The Out-of-Sample Performance of Stochastic Methods in Forecasting Age-Specific Mortality Rates