Point and interval forecasts of age-specific life expectancies

Shang, Han Lin

doi:10.4054/demres.2012.27.21

Cited by 38 publications

(48 citation statements)

References 51 publications

(80 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Smoothing can reduce measurement error and increase the signal-to-noise ratio, and also deals with estimating missing values for some ages at a given year. (4) Given that the functional time-series method can consider more than one component, Shang (2012) indicated that the functional time-series method outperforms the LC method.…”

Section: Functional Time-series Methodsmentioning

confidence: 99%

Dynamic Principal Component Regression: Application to Age-Specific Mortality Forecasting

Shang

2019

ASTIN Bull.

Self Cite

View full text Add to dashboard Cite

In areas of application, including actuarial science and demography, it is increasingly common to consider a time series of curves; an example of this is age-specific mortality rates observed over a period of years. Given that age can be treated as a discrete or continuous variable, a dimension reduction technique, such as principal component analysis, is often implemented. However, in the presence of moderate to strong temporal dependence, static principal component analysis commonly used for analyzing independent and identically distributed data may not be adequate. As an alternative, we consider a dynamic principal component approach to model temporal dependence in a time series of curves. Inspired by Brillinger's (1974) theory of dynamic principal components, we introduce a dynamic principal component analysis, which is based on eigen-decomposition of estimated long-run covariance. Through a series of empirical applications, we demonstrate the potential improvement of one-year-ahead point and interval forecast accuracies that the dynamic principal component regression entails when compared with the static counterpart.

show abstract

Section: Functional Time-series Methodsmentioning

confidence: 99%

Dynamic Principal Component Regression: Application to Age-Specific Mortality Forecasting

Shang

2019

ASTIN Bull.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although the Lee-Carter model has been used and revised extensively since it was first developed in 1992 (Booth and Tickle 2008 ; Booth et al 2006 ; Butt and Haberman 2010 ; Shang 2012 ; Shang et al 2011 ), we use its original version as a benchmark in our case studies. The Lee-Carter model forecasts mortality by age and calendar year on the logarithmic scale while assuming that the relative changes in mortality were constant between the ages over time.…”

Section: Mortality Forecasting Approaches That Tackle Variability Of mentioning

confidence: 99%

Lifespan Disparity as an Additional Indicator for Evaluating Mortality Forecasts

2017

View full text Add to dashboard Cite

Evaluating the predictive ability of mortality forecasts is important yet difficult. Death rates and mean lifespan are basic life table functions typically used to analyze to what extent the forecasts deviate from their realized values. Although these parameters are useful for specifying precisely how mortality has been forecasted, they cannot be used to assess whether the underlying mortality developments are plausible. We therefore propose that in addition to looking at average lifespan, we should examine whether the forecasted variability of the age at death is a plausible continuation of past trends. The validation of mortality forecasts for Italy, Japan, and Denmark demonstrates that their predictive performance can be evaluated more comprehensively by analyzing both the average lifespan and lifespan disparity—that is, by jointly analyzing the mean and the dispersion of mortality. Approaches that account for dynamic age shifts in survival improvements appear to perform better than others that enforce relatively invariant patterns. However, because forecasting approaches are designed to capture trends in average mortality, we argue that studying lifespan disparity may also help to improve the methodology and thus the predictive ability of mortality forecasts.Electronic supplementary materialThe online version of this article (doi:10.1007/s13524-017-0584-0) contains supplementary material, which is available to authorized users.

show abstract

“…Even though the mortality pattern is still different from that of Western European countries (see Vallin and Meslé 2001 ), there are also some similarities ( Bálint and Kovács 2015 ), including increasing population aging (see Gavrilova and Gavrilov 2009 ). Due to limitations of long time series data (with good quality), it is possible that high-mortality regime forecasts will not be the same as those obtained by Shang et al (2011) or Shang (2012) . This expectation is somewhat confirmed by Bohk and Rau (2015) who compared and contrasted forecasting techniques to evaluate the impact of the recent financial crisis on some of the CEE countries and stated that irregular mortality developments are particularly difficult to forecast due to major changes in long-term trends.…”

Section: Introductionmentioning

confidence: 99%

“…As mentioned, LC variants are widely used because of their simplicity; many European countries use LC variants for official forecasts ( Stoeldraijer et al 2013 ). Basic LC method is the main one used for comparisons due to its widespread acceptability (see Booth et al 2012 , Hyndman and Ullah 2007 , Shang 2002 , for example). Moreover, we considered two more approaches for comparison—coherent mortality forecasting and life expectancy forecasting using a Bayesian Hierarchical model adapted by the United Nations (UN).…”

Section: Introductionmentioning

confidence: 99%

Mortality and life expectancy forecast for (comparatively) high mortality countries

Rabbi

Mazzuco

2018

Genus

View full text Add to dashboard Cite

BackgroundThe Lee–Carter method and its later variants are widely accepted extrapolative methods for forecasting mortality and life expectancy in industrial countries due to their simplicity and availability of high quality long time series data.ObjectiveWe compared and contrasted mortality forecasting models for higher mortality regimes that lack long time series data of good quality, which is common in several Central and Eastern European (CEE) countries.Data and methodsWe utilized seven different variants of the Lee–Carter method and coherent mortality forecasts of various CEE countries, and the Bayesian Hierarchical Model used by the United Nations to produce probabilistic forecasts. The data of nine CEE countries with comparatively higher mortality have been considered.ResultsThe performance of the forecasting models for the nine CEE countries was found to be lower than that observed for low-mortality countries. No model gives uniquely best performance for all the nine CEE countries. Most of the LC variants produced lower forecasts of life expectancies than current life expectancy values for Belarus, Russia, and Ukraine. A coherent mortality forecast could not overcome the limitations of single population forecasting techniques due to increasing mortality differences between these countries over the fitting period (mortality divergence). In the same context, the use of the probabilistic forecasting technique from the Bayesian framework resulted in a better forecast than some of the extrapolative methods but also produced a wider prediction interval for several countries. The more detailed analysis for Hungary indicates that a better fit of certain forecasting methods may occur in the later part of the life span rather than the whole life span.ConclusionThese findings imply the necessity of inventing a new forecasting technique for high-mortality countries.

show abstract

Point and interval forecasts of age-specific life expectancies

Cited by 38 publications

References 51 publications

Dynamic Principal Component Regression: Application to Age-Specific Mortality Forecasting

Dynamic Principal Component Regression: Application to Age-Specific Mortality Forecasting

Lifespan Disparity as an Additional Indicator for Evaluating Mortality Forecasts

Mortality and life expectancy forecast for (comparatively) high mortality countries

Contact Info

Product

Resources

About