| c212.LSL {c212} | R Documentation |
Implementaion of the least-slope estimator estimator (LSL) for the proportion of true null hypotheses.
Description
The least-slope estimator estimator (LSL) is one of a number of estimators of the proportion of true null hypotheses. This implementation assumes a grouped structure for the data.
Usage
c212.LSL(trial.data)
Arguments
trial.data |
Data frame containing the p-values for the hypotheses being tested. The data must contain the following columns: B: the index or name of the groupings; p: the p-values of the hypotheses. |
Value
An estimate of the proportion of true null hypotheses.
Note
The implementation is that described in Hu, J. X. and Zhao, H. and Zhou, H. H. (2010).
Author(s)
R. Carragher<raymond.carragher@strath.ac.uk>
References
Hu, J. X. and Zhao, H. and Zhou, H. H. (2010). False Discovery Rate Control With Groups. J Am Stat Assoc, 105(491):1215-1227.
Benjamini Y, Hochberg Y. (2000). On the Adaptive Control of the False Discovery Rate in Multiple Testing With Independent Statistics. Journal of Educational and Behavioral Statistics, 25(1):60–83.
Examples
data(c212.FDR.data)
lsl <- c212.LSL(c212.FDR.data)
print(lsl)
## Not run:
B pi0
1 Bdy-sys_5 1.0000000
2 Bdy-sys_6 1.0000000
3 Bdy-sys_7 1.0000000
4 Bdy-sys_8 1.0000000
5 Bdy-sys_2 1.0000000
6 Bdy-sys_3 0.2857143
7 Bdy-sys_4 1.0000000
8 Bdy-sys_1 1.0000000
## End(Not run)