Figure 3From: A model-based circular binary segmentation algorithm for the analysis of array CGH dataThe GEV distribution models maximal- t distribution adequately. The maximum likelihood GEV distribution fitted maximal-t distribution well. The dashed lines refer to the kernel smoothing density of the maximal-t statistics derived from random numbers, and the solid lines are the maximum likelihood GEV distribution. Three examples under different conditions are shown in the figure. Case 1: number of probes N = 250, skewness sk = 0, kurtosis ku = 3; Case 2: N = 400, sk = 0.5, ku = 4; and Case 3: N = 550, sk = 1, ku = 5.Back to article page