Figure 4
From: Combining test statistics and models in bootstrapped model rejection: it is a balancing act
Type I error rate plots for the single test statistics and their simplistic combinations. Type I error plots show the Expected False Positive Rate (EFPR), which is the stated number of erroneous rejections of a true model, versus the observed number of rejections, Observed False Positive Rate (OFPR). (A) Static example (B) Dynamic examples. As can be seen, only the single test statistics, and none of the combinations, lie along the identity line. The simplistic combinations are therefore disregarded from further analysis.