Gender-specific mortality surfaces are panels of time series of mortality rates that allow to examine the temporal evolution of male and female mortality across ages. The analysis of these surfaces is often complicated by time-varying effects that reflect the association of age and gender with mortality under unobserved time-varying conditions of the population under study. We propose a hidden Markov model as a simple tool to estimate time-varying effects in mortality surfaces. Under this model, age and gender effects depend on the evolution of an unobserved (hidden) Markov chain, which segments each time series of rates according to time-varying latent classes. We describe the details of an efficient EM algorithm for maximum likelihood estimation of the parameters and suggest a straightforward parametric bootstrap routine to compute standard errors. These methods are illustrated on cardiovascular and cancer mortality rates, observed in Italy during the period 1980–2014.
AntonioKBardoutsosAOuburgW (2015) Bayesian Poisson log-bilinear models for mortality projections with multiple populations. European Actuarial Journal, 5, 245–281.
2.
ArthurWBVaupelJW (1984) Some general relationships in population dynamics. Population Index, 50, 214–226.
3.
BarbiECamardaC (2011) Period and cohort effects on elderly mortality: A new relational model for smoothing mortality surfaces. Statistica, 71, 51–69.
4.
BarbiECaselliGYashinA (editors) (2004) Age and time pattern of mortality by cause in Italy: A mortality surface approach. In Proceeding of the XLII scientic meeting of the Italian society of statistics. Padua: CLEUP.
5.
BartolucciFFarcomeniAPennoniF (2012) Latent Markov Models for Longitudinal Data. London: Chapman and Hall.
6.
BiernackiCCeleuxGGovaertG (2000) Assessing a mixture model for clustering with the integrated completed likelihood. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22, 719–725.
7.
BiernackiCCeleuxGGovaertG (2003) Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models. Computational Statistics & Data Analysis, 41, 561–575.
8.
BolanoD (2017) Using hidden Markov models to model life course trajectories. In Proceedings of the Population Association of America (PAA) Annual Meeting, 27–9 April, 2017, Chicago, pages 112.
9.
BrouhnsNDenuitMVermuntJK (2002) A Poisson log-bilinear regression approach to the construction of projected lifetables. Insurance: Mathematics and Economics, 31, 373–393.
10.
BullaJLagonaFMaruottiAPiconeM (2012) A multivariate hidden Markov model for the identification of sea regimes from incomplete skewed and circular time series. Journal of Agricultural, Biological, and Environmental Statistics, 17, 544.
11.
CappéOMoulinesERydénT (2005) Inference in Hidden Markov Models. New York: Springer.
12.
CarstensenB (2007) Age-period-cohort models for the Lexis diagram. Statistics in Medicine, 26, 3018–3045.
13.
CaswellH (2009) Stage, age and individual stochasticity in demography. Oikos, 118, 1763–1782.
14.
ClaytonDSchifflersE (1987) Models for temporal variation in cancer rates. I: Age–period and age–cohort models. Statistics in Medicine, 6, 449–467.
DebnAMontesFMateuJPorcuEBevilacquaM (2008) Modelling residuals dependence in dynamic life tables: A geostatistical approach. Computational Statistics & Data Analysis, 52, 3128–3147.
18.
FreyB (1998) Graphical Models for Machine Learning and Digital Communication. Cambridge, Massachusetts: MIT Press.
19.
LagonaFJdanovDShkolnikovaM (2014) Latent time-varying factors in longitudinal analysis: A linear mixed hidden Markov model for heart rates. Statistics in Medicine, 33, 4116–4134.
20.
LagonaFMaruottiAPadovanoF (2015) Multilevel multivariate modelling of legislative count data, with a hidden Markov chain. Journal of the Royal Statistical Society: Series A (Statistics in Society), 178, 705–723.
LeeRDCarterLR (1992) Modeling and forecasting U.S. mortality. Journal of the American Statistical Association, 87, 659–671.
24.
LiHO'hareCVahidF (2016) Two-dimensional kernel smoothing of mortality surface: An evaluation of cohort strength. Journal of Forecasting, 35, 553–563.
25.
LiNLeeRD (2005) Coherent mortality forecasts for a group of populations: An extension of the Lee–Carter method. Demography, 43, 575–594.
26.
LuomaAPuustelliAKoskinenL (2012) A Bayesian smoothing spline method for mortality modelling. Annals of Actuarial Science, 6, 284–306.
27.
MaruottiA (2011) Mixed hidden Markov models for longitudinal data: An overview. International Statistical Review, 79, 427–454.
28.
MaruottiABullaJLagonaFPiconeMMartellaF (2017) Dynamic mixtures of factor analyzers to characterize multivariate air pollutant exposures. The Annals of Applied Statistics, 11, 1617–1648.
29.
McFarlandDD (1970) Intragenerational social mobility as a Markov process: Including a time-stationary Markovian model that explains observed declines in mobility rates over time. American Sociological Review, 35, 465–43.
30.
McLachlanGPeelD (2000) Finite Mixture Models. New York, NY: Wiley.
31.
PedrozaC (2006) A Bayesian forecasting model: Predicting U.S. male mortality. Biostatistics, 7, 530–550.
32.
PohleJLangrockRvan BeestFMSchmidtNM (2017) Selecting the number of states in hidden Markov models: Pragmatic solutions illustrated using animal movement. Journal of Agricultural, Biological and Environmental Statistics, 22, 270–293.
33.
RenshawAHabermanS (2003) On the forecasting of mortality reduction factors. Insurance: Mathematics and Economics, 32, 379–401.
34.
RenshawAHabermanS (2006) A cohort-based extension to the Lee–Carter model for mortality reduction factors. Insurance: Mathematics and Economics, 38, 556–570.
35.
VerbekeGMolenberghsG (2000) Linear Mixed Models for Longitudinal Data. New York: Springer.
36.
VisserIRaijmakersMMolenaarP (2000) Confidence intervals for hidden Markov model parameters. British Journal of Mathematical and Statistical Psychology, 53, 317–327.
37.
WangPCockburnIPutermanM (1998) Analysis of patent data: A mixed-Poisson-regression-model approach. Journal of Business and Economic Statistics, 16, 27–41.
38.
WangPPutermanML (1999) Markov Poisson regression models for discrete time series. Part 1: Methodology. Journal of Applied Statistics, 26, 855–869.
39.
WedelMDesarboWSBultJRRamaswamyV (1993) A latent class Poisson regression model for heterogeneous count data. Journal of Applied Econometrics, 8, 397–411.
40.
WillekensF (2014) Multistate Analysis of Life Histories with R. Berlin: Springer-Verlag.
41.
WilmothJR (1990) Variation in vital rates by age, period, and cohort. Sociological Methodology, 20, 295–335.
42.
WilmothJVallinJCaselliG (1989) Quand certaines gnrations ont une mortalit dirente de celle que l'on pourrait attendre [Differences between observed and expected generation mortality]. Population, 44, 335–376.
43.
WisniowskiASmithPWFBijakJRaymerJForsterJJ (2015) Bayesian population forecasting: Extending the Lee–Carter method. Demography, 52, 1035–1059.
44.
WuC (1983) On the convergence properties of the EM algorithm. The Annals of Statistics, 11, 95–103.
45.
ZucchiniWMacDonaldILLangrockR (2016) Hidden Markov Models for Time Series. An Introduction Using R, 2nd edition. Boca Raton, FL: CRC Press.
