#### Probabilistic forecasting of maximum human lifespan by 2100 using Bayesian population projections - Replicability Files
#### Authors: Michael Pearce and Adrian E. Raftery
#### Last major modification: February 8, 2021
#### Software Version: R 3.6.1

#### Files
(1) 5034_code.R contains the code used to run analyses and create figures/tables for the paper.


#### IDL Data
The original data from IDL is not included by the authors, as it is not public domain. It can be downloaded from https://www.supercentenarians.org/ after creating a free account and agreeing to their terms and conditions. Below is a brief meta-summary of the data:

Total Number of Observations: 15,050 (each observation is a verified semi- or super-centenarian death record)
Total Number of Variables: 24
Variables Used in Analysis: "AGEYEARS" [age at death in years], "DAYSSINCEBD" [days since last birthday at death], "AGEDAYS" [age at death in days], "SEX" [M/F], "DCOUNTRY" [country of death], "BDATE" [birth date], "DDATE" [death date], "UPDATE" [IDL update that first included observation; {1,2,3}]  

#### bayesPop Projections Data
This data should be a folder, called mcmc, which contains all total fertility rate (TFR) and life expectancy at birth (e0) projections to be used by the bayesPop package when creating probabilistic population projections. As stated in the 5034_code.R file, this data can be found directly directly from the bayesPop website, https://bayespop.csss.washington.edu/software/. It is not included here simply due to its very large size.