Figure 7

Comparison of statistical power of classifiers when feature distributions within class are Gaussian and the biomarkers are correlated. Each dataset included 1000 features per subject, where features were distributed according to a Gaussian distribution within each class. The results are based on k= 1% and n= 150. In order to simulate dependence between biomarkers, the data generation model assumed a correlation of 0.5 between the 10 biomarkers in both cases and controls. Results are based on 100 simulated datasets.