parameter_mapping: Mapping parameters to the model

In MJomaba/flu-evidence-synthesis: Flu Evidence Synthesis: Analysing Effectiveness of Vaccine Strategies

Mapping parameters to the model

Description

To reduce complexity it is common to map certain parameters to multiple age groups. For example in the UK model we use the same susceptibility for age groups 1,2 and 3. Similarly for fitting purposes we need a vector of parameters. That vector will hold different parameters: ascertainment (epsilon; by age group), psi, transmissibility, susceptibility (by age group) and initial infected. This method makes this easier.

In the UK model we use 9 parameters, the first three are the epsilon for age groups 1,2 and 3,4 and 5. This is then followed by the psi parameter (4th parameter in the parameter vector). Next is the transmissibility parameter. Then three susceptibility parameters, used in age groups: 1,2,3 and 4,5,6 and 7. Finally followed by initial_infected as the 9th parameter in the parameters vector. parameter_map(epsilon = c(1,1,2,2,3), psi = 4, transmissibility = 5, susceptibility = c(6,6,6,7,7,7,8), initial_infected = 9

Usage

parameter_mapping(epsilon, psi, transmissibility, susceptibility,
  initial_infected, parameters)

Arguments

epsilon	A vector holding the indeces of the ascertainment parameters for each age group in the parameters vector
psi	Index of the psi parameter in the parameters vector
transmissibility	Index of the transmissibility parameter in the parameters vector
susceptibility	A vector holding the indeces of the susceptibility parameter for each age group in the parameters vector
initial_infected	Index of the initial_infected parameter in the parameters vector
parameters	Optional parameters vector. For simple cases it is possible to infer the parameter map just from the (named) parameters vector. In that case no other variables need to be passed to this function.

Value

A list specifying the parameter mapping as used in inference and vaccination_scenario

MJomaba/flu-evidence-synthesis documentation built on April 26, 2022, 11:12 p.m.