R/final_contrasts.R

In fdrennan/test: What the Package Does (One Line, Title Case)

# This function estimates the adjusted confidence intervals and p-values based
# on a single set of tests. Here the adjust = 'mvt' corresponds to the simulate
# option in proc glimmix SAS which simulates potential test statistics from a
# multivariate t-distribution with accounts for the correlation and the degrees
# of freedom.

final_contrasts <- function(model, cont_list, est) {
  if (length(cont_list) > 0) {
    # Compute the simulated p-value
    adj_ht <- contrast(est, adjust = "mvt", method = cont_list) %>% data.frame()

    # Compute confidence interval for the the contrast
    adj_ci <- confint(contrast(est, adjust = "mvt", method = cont_list)) %>% data.frame()

    # Some rounding, this can be done here or when results are output in excel
    # It might make for sense to do it later, if we are able to use the scientific
    # notation in excel

    results <- inner_join(adj_ht, adj_ci) %>%
      mutate(
        estimate = round(estimate, 2),
        SE = round(SE, 3),
        df = round(df, 2),
        t.ratio = round(t.ratio, 2),
        p.value = round(p.value, 3),
        lower.CL = round(lower.CL, 3),
        upper.CL = round(upper.CL, 3)
      )
    return(results)
  }
}