Bernd Heidergott (Tinbergen Institute & Free University Amsterdam)
A Network View on Aging and Mortality: the Gompertz- Makeham law for mortality
The mortality rate m(t) at age t describes the probability that a person dies in a short rage interval following age t. Mitnitski et al. introduced a dynamic network for modelling the mortality rate where nodes represent the state of health indicators (healthy or damaged), the dynamic relationship between the states of the nodes models the interdependency of the health indicators (damaged can spread through the network), and death is defined as the time it takes to reach a network state in which two carefully selected nodes, called mortality nodes, are both damaged. The intriguing aspect of the Mitnitski mortality network model is that it works very well : -) but so far is lacking a mathematically rigorous underpinning : -(
In our talk, we will give a detailed introduction of the Mitnitski model use for modelling of mortality rate (as far as we understand it). Our numerical experiments show a good fit to actual mortality rate data, and we discuss interpretation of death as the accumulation of damaged health indicators.
We are still working on a deeper understanding of the network process and in our talk is more of travelogue journal than an account of what to find at the final destination.
joint work with Valentin Flietner, Frank den Hollander, Ines Linder, and Holger Struklik