bp_loglike_binary_cov: Calculate the likelihood for binary covariate MC trees
In skgallagher/InfectionTrees: Sample, Analyze, and Plot Infection Trees

Description Usage Arguments Details Value Examples

Calculate the likelihood for binary covariate MC trees

bp_loglike_binary_cov(
  inf_params,
  obs_data_summary,
  mc_samples_summary,
  return_neg = FALSE,
  messages = FALSE
)

`inf_params`	vector of length 2, the beta0 and beta1 coefficients
`obs_data_summary`	cluster_size number in cluster x_pos number of positive smears in cluster x_neg number of negative smears in the cluster freq number of times this frequency occurs
`mc_samples_summary`	a data frame of sampled transmission trees with the following columns, x_pos_trans number of total transmissions from any smear positive person x_pos number of positive smears x_neg number of negative smears cluster_size cluster size mc_freq number of samples
`return_neg`	logical indicating whether to return the negative loglike. Default is FALSE
`messages`	logical indicating whether we should print messages. Default is FALSE.

WARNING: obs_data_summary and mc_samples_summary must not both have a 'freq' column.

the log average likelihood over the observed clusters

library(dplyr)
inf_params <- c(-1, 1)
mc_samples_summary <- data.frame(freq = c(10, 5, 6, 1, 1),
                                mc_freq = c(10, 10, 10, 3, 7),
                                cluster_size = c(1, 1, 2, 2, 2),
                                x_pos = c(1, 0, 2, 1, 1),
                                x_neg = c(0, 1, 0, 1, 1),
                                x_pos_trans = c(0, 0, 1, 1, 0),
                                x_neg_trans = c(0, 0, 0, 0, 1))
obs_data_summary <- mc_samples_summary %>%
 group_by(cluster_size, x_pos, x_neg) %>%
 summarize(freq = sum(freq), .groups = "drop")
bp_loglike_binary_cov(inf_params = inf_params,
obs_data_summary = obs_data_summary,
mc_samples_summary = mc_samples_summary %>%
  dplyr::select(-freq),
return_neg = FALSE)