Dynamic modelling of mortality via mixtures of skewed distribution functions

There has been growing interest on forecasting mortality. In this article, we propose a novel dynamic Bayesian approach for modeling and forecasting the age-at-death distribution, focusing on a three-components mixture of a Dirac mass, a Gaussian distribution and a Skew-Normal distribution. According to the specified model, the age-at-death distribution is characterized via seven parameters corresponding to the main aspects of infant, adult and old-age mortality. The proposed approach focuses on coherent modeling of multiple countries, and following a Bayesian approach to inference we allow to borrow information across populations and to shrink parameters towards a common mean level, implicitly penalizing diverging scenarios. Dynamic modeling across years is induced through an hierarchical dynamic prior distribution that allows to characterize the temporal evolution of each mortality component and to forecast the age-at-death distribution. Empirical results on multiple countries indicate that the proposed approach outperforms popular methods for forecasting mortality, providing interpretable insights on its evolution.

[1]  Michael A. West,et al.  Time Series: Modeling, Computation, and Inference , 2010 .

[2]  Marius D. Pascariu,et al.  The maximum entropy mortality model: forecasting mortality using statistical moments , 2019, Scandinavian Actuarial Journal.

[3]  Rob J. Hyndman,et al.  Robust forecasting of mortality and fertility rates: A functional data approach , 2007, Comput. Stat. Data Anal..

[4]  S. Mazzuco,et al.  A Mixture-Function Mortality Model: Illustration of the Evolution of Premature Mortality , 2020, European journal of population = Revue europeenne de demographie.

[5]  Thomas B. L. Kirkwood,et al.  Deciphering death: a commentary on Gompertz (1825) ‘On the nature of the function expressive of the law of human mortality, and on a new mode of determining the value of life contingencies’ , 2015, Philosophical Transactions of the Royal Society B: Biological Sciences.

[6]  D. Owen.  Tables for Computing Bivariate Normal Probabilities , 1956 .

[7]  Melvin J. Hinich,et al.  Time Series Analysis by State Space Methods , 2001 .

[8]  Wolfgang Lutz,et al.  Dimensions of global population projections: what do we know about future population trends and structures? , 2010, Philosophical Transactions of the Royal Society B: Biological Sciences.

[9]  J. Robine,et al.  Three dimensions of the survival curve: horizontalization, verticalization, and longevity extension , 2005, Demography.

[10]  M. Newton,et al.  Estimating the Integrated Likelihood via Posterior Simulation Using the Harmonic Mean Identity , 2006 .

[11]  V. Canudas-Romo.  Three measures of longevity: Time trends and record values , 2010, Demography.

[12]  Jun Yan.  Survival Analysis: Techniques for Censored and Truncated Data , 2004 .

[13]  H. Haario,et al.  An adaptive Metropolis algorithm , 2001 .

[14]  H. Booth.  Coherent Mortality Forecasting with Standards: Low Mortality Serves as a Guide , 2020, Developments in Demographic Forecasting.

[15]  Petros Dellaportas,et al.  Bayesian analysis of mortality data , 2001 .

[16]  J. McKeen Cattell,et al.  The Chances of Death and other Studies in Evolution , 1897 .

[17]  Marius D. Pascariu,et al.  Alternative Forecasts of Danish Life Expectancy , 2020, Developments in Demographic Forecasting.

[18]  Chris Wilson,et al.  On the scale of global demographic convergence, 1950-2000. , 2001, Population and development review.

[19]  C. Finch,et al.  The rise and fall of excess male infant mortality , 2008, Proceedings of the National Academy of Sciences.

[20]  Nan Li,et al.  Coherent mortality forecasts for a group of populations: An extension of the lee-carter method , 2005, Demography.

[21]  Bruce D. McCullough,et al.  Statistical Analysis of Stochastic Processes in Time , 2005, Technometrics.

[22]  Jackie Li.  A Poisson common factor model for projecting mortality and life expectancy jointly for females and males , 2013, Population studies.

[23]  J. H. Pollard,et al.  The age pattern of mortality , 1979 .

[24]  M. Marmot,et al.  Premature mortality attributable to socioeconomic inequality in England between 2003 and 2018: an observational study , 2019, The Lancet. Public health.

[25]  Nicolas Chopin,et al.  An Introduction to Sequential Monte Carlo , 2020 .

[26]  J. Oeppen,et al.  Coherent forecasting of multiple-decrement life tables: a test using Japanese cause of death data , 2008 .

[27]  C. Camarda,et al.  Modelling and forecasting adult age-at-death distributions , 2019, Population studies.

[28]  Adelchi Azzalini,et al.  The Skew-Normal and Related Families , 2018 .

[29]  Ugofilippo Basellini,et al.  A Three-Component Approach to Model and Forecast Age-at-Death Distributions , 2020, Developments in Demographic Forecasting.

[30]  Rob J Hyndman,et al.  Coherent Mortality Forecasting: The Product-Ratio Method With Functional Time Series Models , 2013, Demography.

[31]  S. Mazzuco,et al.  A mortality model based on a mixture distribution function , 2018, Population studies.

[32]  Ute Beyer,et al.  Bayesian Forecasting And Dynamic Models , 2016 .

[33]  Kaare Christensen,et al.  Rise, stagnation, and rise of Danish women’s life expectancy , 2016, Proceedings of the National Academy of Sciences.

[34]  J. Vaupel,et al.  The importance of the reference populations for coherent mortality forecasting models , 2016 .

[35]  Nathan Keyfitz,et al.  Applied Mathematical Demography , 1978 .

[36]  Jacques F. Carriere,et al.  PARAMETRIC MODELS FOR LIFE TABLES , 1992 .

[37]  W. M. Makeham,et al.  On the Law of Mortality and the Construction of Annuity Tables , 1860 .

[38]  W Siler,et al.  Parameters of mortality in human populations with widely varying life spans. , 2007, Statistics in medicine.

[39]  J. Vaupel,et al.  Coherent Forecasts of Mortality with Compositional Data Analysis , 2017 .