multiloglikechain

pars

a collection of <code>[(2+nc) x m ]</code> matrices comprised of outcomes (first row), treatments (second row), and confounders (from the third row), where <code>nc</code> is the number of confounders.

listobservations

a matrix comprised of every possible values for outcome in each row.

permutetab

a matrix of which each row indicates a pair of index for adjacent outcomes.

edgeY

a matrix of which each row indicates a index for treatment (first column) and for outcome (second column) on which the treatment has a direct effect.

edgeAY

a list of edges of which a list of matrix specifying additional directed edges (from confounders or treatment to the outcomes) information.

edgeExtra

Derive log-likelihood of conditional log-linear model given parameters.

In networks, treatments may spill over from the treated individual to his or her social contacts and outcomes may be contagious over time. Under this setting, causal inference on the collective outcome observed over all network is often of interest. We use chain graph models approximating the projection of the full longitudinal data onto the observed data to identify the causal effect of the intervention on the whole outcome. Justification of such approximation is demonstrated in Ogburn et al. (2018) <arXiv:1812.04990>.

multiloglikechain: Derive log-likelihood of conditional log-linear model given parameters.

Description

Usage

Arguments

Value