Alternative HCT Procedure to Choose P-Value Threshold Based on Beta Distribution of P-Values.
This procedure chooses the p-value threshold for feature selection in a similar fashion to hct_empirical. However, it is based on the Beta distribution of the p-values. Only the features whose p-values are less than the thresold will be included in the classifier.
hct_beta(pvalue, p, n)
A vector containing the p*alpha_0 smallest p-values; typically alpha_0=0.10
The number of the features in total.
The total sample size for the two groups.
Refer to Sanchez, et al (2016), Section 3 and supplementary materials.
The p-value threshold for feature selection. Only the features whose p-values are less than the threshold will be included in the classifier.
Meihua Wu <firstname.lastname@example.org> Brisa N. Sanchez <email@example.com> Peter X.K. Song <firstname.lastname@example.org> Raymond Luu <email@example.com> Wen Wang <firstname.lastname@example.org>
Sanchez, B.N., Wu, M., Song, P.X.K., and Wang W. (2016). "Study design in high-dimensional classification analysis." Biostatistics, in press.
hct_beta(pvalue=0.10,p=500,n=80) # 0.1