Biology Reference
In-Depth Information
t ¼ lnð1 b
3
lnðUÞ=a
3
Þ=b
3
: (11.16)
Multiple component models do not generally have explicit solutions, but one can solve the
survivorship numerically using “uniroot” in “R.”
Figure 11.5
shows the results of simulating
10,000 deaths from a combined negative and positive Gompertz model that
Nagaoka et al.
(2006)
fit to a Medieval Japanese archaeological skeletal collection. For completeness,
Figure 11.5
also shows the hazard function, the survivorship function, and the age-at-death
distribution from the hazard model.
Survivorship
Hazard
0
20
40
60
80
0
20
40
60
80
Age
Age
Age−at−Death
Simulated Ages
0
20
40
60
80
0
20
40
60
80
Age
Ages
FIGURE 11.5
Hazard function, survivorship function, and probability density function (for age-at-death) from
a negative and positive Gompertz model taken from
Nagaoka et al.'s (2006)
analysis of a Medieval Japanese
archaeological skeletal collection. The final panel shows a simulation of 10,000 deaths from the modeled
distribution.