R/log_norm.R

In cnfoley/mrclust: Identifying Clustered Heterogeneity in Mendelian Randomization Analyses

log_norm <- function(i, j, tmp_lgt, tmp_lgt2, theta, theta_sd, theta_clust) {
  tmp_theta <- rep(theta[i], tmp_lgt2)
  tmp_theta_sd <- rep(theta_sd[i], tmp_lgt2)
  tmp_clust <- rep(theta_clust[j], each = tmp_lgt)
  res <- (-0.5 * log(2 * pi) - log(tmp_theta_sd)
          - 0.5 * (tmp_theta - tmp_clust)^2 / tmp_theta_sd^2)
  return(res)
}