War Premium Over? from Kim Zussman
Using Wiki data there does appear to be a reduction in % of world population killed in war over time.
http://en.wikipedia.org/wiki/List_of_wars_and_anthropogenic_disasters_by_death_toll
The high and low death toll estimates were averaged for the dependent variable, and the mid-point year of the war period was used as the independent variable. Regressing mean estimated % of world population killed in war vs year:
The regression equation is
mean est % world pop killed = 9.94 - 0.00466 avg year
Predictor Coef SE Coef T P
Constant 9.938 2.693 3.69 0.001
avg year -0.005 0.0016 -2.97 0.007
S = 3.26307 R-Sq = 29.6% R-Sq(adj) = 26.2%
Significant; year explains 26% of the variance. However a plot suggests though war has become less deadly over time, it is more frequent.
