agcEffect.RdagcEffect takes a covariate and outcome model fit
from agcParam, parameters for an MCMC, and the
specified treatment effect and determines the causal estimates (direct and spillover)
generated from the network structure.
agcEffect(mod, burnin = 1, thin = 0.5, treatment_allocation = 0.5, subset = 0, R = 10, burnin_R = 10, burnin_cov = 10, average = TRUE, index_override = 0) # S4 method for list agcEffect(mod, burnin = 1, thin = 0.2, treatment_allocation = 0.5, subset = 0, R = 10, burnin_R = 10, burnin_cov = 10, average = TRUE, index_override = 0)
| mod | An |
|---|---|
| burnin | The index to start evaluation as one would normally have for a burnin for a Bayesian computation. Default = 1 |
| thin | The rate at which to evaluate the outcome model. The closer to 1, the more values to compute (supplying 1 will not thin at all). |
| treatment_allocation | The proportion of the individuals in the network to receive the treatment. Should be a number between 0 and 1. Default = 0.5. |
| subset | The indices of the individuals, as they appear in the adjacency matrix, to be included in the network causal effects estimates. Default = 0 (include everyone). |
| R | The number of iterations for the Gibbs inner loop. Default = 10. |
| burnin_R | The index to start evaluation as one would normally have for a burnin for a Bayesian computation. Default = 10. |
| burnin_cov | The index to start saving covariates for use in the network causal effect chains. Default = 10. |
| average | An indicator of whether to evaluate the causal effects as an average of the R iterations. Default = "TRUE". |
| index_override | The MCMC outer loop iteration numbers to include in the evaluation of the causal effects. This will override the burnin and thin parameters. Default = 0 (all iterations included). |
An S3 object of type agcEffectClass that contains essential
values for the outcome model.
rdsIn <- readRDS(paste0(system.file('extdata',package='autognet'),"/agc-example.rds")) adjmat <- rdsIn[[1]] data <- rdsIn[[2]] treatment <- "treatment" outcome <- "outcome" B = 10 R = 5 seed = 1 mod <- agcParam(data, treatment, outcome, adjmat, B = B, R = R, seed = c(1))#> | | | 0% | |======== | 11% | |================ | 22% | |======================= | 33% | |=============================== | 44% | |======================================= | 56% | |=============================================== | 67% | |====================================================== | 78% | |============================================================== | 89% | |======================================================================| 100%effects <- agcEffect(mod)#> | | | 0% | |======================================================================| 100%