The zip archive contains original input data from the Human Mortality Database, data created as part of the analysis, and R programs that replicate the analysis and graphics. Note that several of the programs use random seeds, so when rerun the output may differ slightly from that reported in the paper. 

The R programs use the "tidyverse" package extensively, and were run using R Version 3.5.3. 

When unzipped the archive will have subdirectors, named as follows
	code/			[R programs for analysis and plots]
	data/			[intermediate data created as part of the analysis]
	plots/			[plots included in the manuscript, in both .pdf and .eps formats]

The HMD files used in the analysis are available in a dedicated replication file,
  https://www.mortality.org/hmd/REPLICATION/Schmertmann_2021_Demog_Res_44-45_data.zip 
that is hosted on the HMD site. Accessing that file, either directly via the web or indirectly in the STEP-0 
script file, requires (free) user registration for the HMD site at
  https://mortality.org

The main programs for replicating the paper, which should be run in order, are 
        STEP-0-download-HMD-replication-data.R
	STEP-1-assemble-female-1x10-data.R
	STEP-2-BIG-cross-validated-HMD-test.R
	STEP-3-summarize-BIG-cross-validated-fits.R