Great article, and as he mentions in a footnote, he doesn't know of any other studies, besides a Cooley Hubbard Study which used Monte Carlo analysis (fake history with no regression to mean in the data series, in other words), which looks at international diversification's effect on portfolio survivability frrom withdrawals.
So this makes me feel good, even though it shows how 60% stocks fare much better than 40% stocks over 40 years (as in the graph above) even though at 30 years the 40% stock portfolios look essentially the same as the 60% ones. which is to say, terrific (Table 2 in the paper).
Now to hunt down some of these new international data series, then try to get some value and small tilts figured into them....
Thanks, Ronin!