Bayesian inference to estimate mortality from incomplete historical data

We present a Bayesian model to estimate adult mortality from historical datasets in which the ages of individuals are unknown. The motivation behind this model is to analyze data from the Barcelona Historical Marriage Database (BHMD), a dataset composed by marriage records in which the ages are not recorded, though some lower and upper times of birth and death can be extracted for many individuals. Here we test the model by applying it to a different historical database from which the exact times of birth and death are known, but that we intentionally “corrupt” in order to make it look like the data from the BHMD. The model uses a hierarchical Bayesian approach that requires the implementation of a Markov chain Monte Carlo (MCMC) algorithm to iteratively estimate mortality parameters and times of birth and death. Preliminary results show that the model is able to recover good estimates of the life expectancy and the ages at death. ∗villavicencio@imada.sdu.dk 1

[1]  Francisco Villavicencio,et al.  Reconstructing Lifespans Through Historical Marriage Records of Barcelona from the Sixteenth and Seventeenth Centuries , 2015, Population Reconstruction.

[2]  J. Gaudemet.  Le mariage en Occident : les mœurs et le droit , 1987 .

[3]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[4]  David P. Smith,et al.  On the Law of Mortality , 2013 .

[5]  Gaspar Feliu i Montfort.  La demografia baixmedieval catalana: estat de la qüestió i propostes de futur , 1999 .

[6]  T. Trussell.  A refined estimator of measures of location of the age at first marriage , 2011, Demography.

[7]  James S. Clark,et al.  Bayesian inference on age-specific survival for censored and truncated data. , 2012, The Journal of animal ecology.

[8]  Jutta Gampe,et al.  Modelling general patterns of digit preference , 2008 .

[9]  Carreras i Candi Francesc.  Les obres de la Catedral de Barcelona 1298-1445 (Acabament) , 1914 .

[10]  D. Mcneil,et al.  The distribution by age of the frequency of first marriage in a female cohort. , 1972, Journal of the American Statistical Association.

[11]  E. J. Thompson.  Regional Model Life Tables and Stable Populations , 1967 .

[12]  O. Jones,et al.  BaSTA: an R package for Bayesian estimation of age‐specific survival from incomplete mark–recapture/recovery data with covariates , 2012 .

[13]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[14]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[15]  S de Lusignan,et al.  End-digit preference in blood pressure recordings of patients with ischaemic heart disease in primary care , 2004, Journal of Human Hypertension.

[16]  Jordi Nadal i Lorenzo,et al.  Immigració i redreç demogràfic: els francesos a la Catalunya dels segles XVI i XVII , 2000 .

[17]  James W. Vaupel,et al.  The force of mortality at ages 80 to 120 , 1998 .

[18]  R. Rowland,et al.  Demografia del censo de Floridablanca. Una aproximacion , 1990, Revista de Historia Económica / Journal of Iberian and Latin American Economic History.

[19]  K. Lynch.  Life Under Pressure: Mortality and Living Standards in Europe and Asia, 1700-1900 , 2005 .

[20]  M. Rowland,et al.  Self-reported weight and height. , 1990, The American journal of clinical nutrition.

[21]  J. Landers.  Life under Pressure: Mortality and Living Standards in Europe and Asia. Edited by Tommy Bengtsson, Cameron Campbell, and James Z. Lee. Cambridge, MA and London: The MIT Press, 2004. Pp. xiv, 531. $45.00. , 2005, The Journal of Economic History.