sir_forward_backward: Forward backward algorithm for the SIR model

In nianqiaoju/agents: Toolkit for inference in agent-based modelss

Description

Forward backward algorithm for the SIR model

Usage

sir_forward_backward(y, model_config, num_samples = 1, all_xx = NULL)

Arguments

y	population observations
model_config	a list containing 'N' : population size 'alpha0' 'lambda' 'gamma' 'rho' 'adjacency_matrix_b': network structure
num_samples	number of backward samples needed, set to 1 if not specifed. If num.samples = 0, then no backward sample is computed.
all_xx	a 2**N by N matrix, representing the sample space, pre-calculated to save time

Value

a list containing

• 'log marginal likelihood and one backward sample'

• 'bs' : backward samples in a [N * num_observations * num_samples] array

nianqiaoju/agents documentation built on Feb. 19, 2021, 12:18 a.m.