R/max_min_assoc.gee.R

In MXM: Feature Selection (Including Multiple Solutions) and Bayesian Networks

max_min_assoc.gee = function(target, reps, group, dataset, test, wei, threshold, max_k, selectedVars, pvalues, stats, remainingVars, 
                             univariateModels, selectedVarsOrder, hash, stat_hash, pvalue_hash, correl, se) {
  selected_var = -1;
  selected_pvalue = 2;
  selected_stat = 0;
  varsToIterate = which(remainingVars==1);
  for (cvar in varsToIterate) {
    mma_res = min_assoc.gee(target, reps, group, dataset, test, max_k, cvar, wei, selectedVars, pvalues, stats, univariateModels, selectedVarsOrder, hash, stat_hash, pvalue_hash, correl = correl, se = se);
    pvalues = mma_res$pvalues;
    stats = mma_res$stats;
    stat_hash = mma_res$stat_hash;
    pvalue_hash = mma_res$pvalue_hash;
    
    if(mma_res$pvalue > threshold)   remainingVars[[cvar]] = 0;
    
    if ( compare_p_values(mma_res$pvalue, selected_pvalue, mma_res$stat, selected_stat) ) {
      selected_var = cvar;
      selected_pvalue = mma_res$pvalue;
      selected_stat = mma_res$stat;
    }
  }
  results <- list(selected_var = selected_var, selected_pvalue = selected_pvalue, remainingVars = remainingVars, pvalues = pvalues, stats = stats, stat_hash=stat_hash, pvalue_hash = pvalue_hash, correl = correl, se = se);
  return(results); 
}