Epidemic risk assessment from geographic population density

Applied Network Science

Table 1 The interpersonal contact models used throughout the paper

Model		Contact probability \(p_{u,v}\)
Name	Acronym
Homogeneous mixing	\(\text {HM}\)	constant
Distance–based mixing	\(\text {DM}\)	\(\propto d_{u,v}^{-1}\)
Distance and fitness–based mixing	\(\text {DFM}\)	\(\propto d_{u,v}^{-1}f_uf_v\)
Distance and age–based mixing	\(\text {DAM}\)	\(\propto s_{g(u),g(v)}d_{u,v}^{-1}\)
Multi–layer mixing	\(\text {MLM}\)	\(=1\) if \((u,v)\in E_H\), \(=0.5\) if \((u,v)\in E_A\),
		\(\propto s_{g(u),g(v)}d_{u,v}^{-1}f_u,f_v\) otherwise

\(p_{u,v}\) is the probability that u and v are connected at each step of the simulation
\(f_u\) is the Lognormally distributed social fitness of u
\(s_{g_u,g_v}\) is the data–driven age–based social mixing for u and v’s age groups \(g_u\) and \(g_v\)
\(E_H\) and \(E_A\) are the edge sets of two static networks, respectively used to represent households and acquaintances (see Sect. 4 and Guarino et al. (2021) for more details)