Skip to main content
Figure 2 | BMC Systems Biology

Figure 2

From: Combining test statistics and models in bootstrapped model rejection: it is a balancing act

Figure 2

A graphical summary of the proposed method steps for a 2D χ 2 vs DW analysis. Encircled numbers correspond to the steps described in the model algorithm. The starting point is some data set,, a model structure, M 1 , to be investigated, and two test statistics. First the model is fitted to the experimental data and then the fitted model is used to generate bootstrap samples. Next, the model is fitted to all bootstrap samples. The resulting two-dimensional cloud is used to generate a density estimation. The cutoff plane is defined as the equidensity contour that goes through the coordinates of the experimental data (red square). The corresponding p-value is equal to the portion of the distribution below the plane, i.e. outside the corresponding density contour line. The p-value is then used for decision on whether or not to reject either model.

Back to article page