Link to the PDF presentation:
To reproduce the analysis and slides follow these steps.
(1) Install the yada package (github: SlothOfDoom/yada) using devtools. Instructions are here:
https://www.michaelholtonprice.com/yada
(2) Clone the saa2018 repository hosted on bitbucket.
git clone https://mpatmudd@bitbucket.org/mpatmudd/saa2018.git
(3) Start R in the directory where saa2018 was cloned. For example, I clone to the folder /bitbucket off my home directory and hence on Linux do the following in a terminal window:
cd ~/bitbucket/saa2018
R
(4) Run the simulation. This can take some time. If you run out of memory, reduce Nmc, the number of Monte Carlo samples for the integration of the posterior.
source("saa2018_simulation.R")
(5) Generate the plots illustrating the Bayesian updating:
source("saa2018_bayesian_update_plots.R")
(6) Compile the beamer presentation. For example, on Linux open a terminal window and run the following in the saa2018 directory:
pdflatex saa2018_mhprice_fully_bayesian_reconstruction_of_past_demography.tex
To ensure identical results, these are the final commits for yada and saa2018 used to generate the SAA 2018 presentation slides.
Latest commit for yada [github]:
0f3fa648dcec114c9fbeace95d94181d08b34366
Latest commit for saa2018 [bitbucket]:
2399b5fce4bef8250cfddf013832dc342b986190