R/ComputeZscores.R

In DCAN-Labs/MarginalModelCIFTI: MarginalModelCifti constructs and tests marginal models for surface and volume data.

#' ComputeZscores -- convert estimated Wald statistics to z scores
#' 
#' This function will compute z scores from Wald statistics, which can be used to infer statistical significance.
#' The function can be called in parallel to operate on a map. Such usage can be found in the main function
#  or in ComputeMM_WB.R. See below for examples.
#' @param geeglm_obj An GLM object that is produced from computing the marginal model
#' @keywords normalization
#' @export
#' @examples 
#' zscores <- ComputeZscores(geeglm_obj)
ComputeZscores <- function(geeglm_obj,nmaps) {
  if (is.numeric(unlist(geeglm_obj))) {
    zscores_model = numeric(nmaps)
  } else
  {
    if (is.object(geeglm_obj)) {
      gee_obj = geeglm_obj
    } else
    {
      gee_obj <- geeglm_obj$V
    }
    zscores_model <- qnorm(0.99999999999999-pchisq((coef(gee_obj)/sqrt(diag(gee_obj$geese$vbeta)))^2,1, lower.tail=F))    
  }
  return(zscores_model)
}