Skip to main content

Table 6 Relative bias in the two-component mixture model

From: Segmentation and intensity estimation for microarray images with saturated pixels

Parameter GMM0 CGMM GMM1
  10 40 70 10 40 70 10 40 70
π1 = 0.8 0.0006 0.0007 0.0006 0.0006 0.0006 0.0006 0.0026 0.0075 0.0035
μ1 = 8, 000 0.0003 0.0003 0.0003 0.0003 0.0003 0.0003 0.0007 0.0006 0.0014
σ1 = 2, 000 -0.0043 -0.0043 -0.0044 -0.0044 -0.0044 -0.0045 -0.0010 0.0032 0.0073
μ 2 -0.0002 0.0000 -0.0001 -0.0001 0.0004 -0.0019 -0.0104 -0.0867 -0.2220
σ 2 -0.0079 -0.0080 -0.0075 -0.0070 -0.0068 -0.0162 -0.1121 -0.3784 -0.6426
  1. Simulation with true K = 2: Relative bias based on runs with K correctly selected by BIC. The models considered were regular Gaussian mixture for complete, uncensored data (GMM0), censored Gaussian mixture for censored data, and regular Gaussian mixture for censored data (GMM1). Percents of saturated foreground pixels were set at 10% (μ2 = 46, 300, σ2 = 15, 000), 40% (μ2 = 60, 450, σ2 = 20, 000) and 70% (μ2 = 78, 650, σ2 = 25, 000).