Description Usage Arguments Value Author(s)
moderated SE's are fit$stdev.unscaled * sqrt(fit$s2.post) unmoderated SE's are fit$stdev.unscaled * fit$sigma from p53 & 54 limma userguide: for gene j, contrast k: ujk = fit$stdev.unscaled = unscaled stdev sj = fit$sigma = gene j's residual variance, with dj df Bjk = linear model estimates = the logFC's tstat.ord <- signal/SE <- Bjk/(ujk.sj) <- fit$coef/(fit$stdev.unscaled * fit$sigma) tstat.mod <- signal/SE' <- Bjk/(ujk.s'j) <- fit$coef/(fit$stdev.unscaled * sqrt(fit$s2.post)) ^^^^ which follows a t-dist with d0 + dj d.f. s2.post is the weighted avg of s2.prior and sigma^2 with weights proportional to df.prior and df.residual, respectively.
1 | topTable.SE(fit, coef = NULL, ...)
|
fit |
see topTable |
coef |
the coefficient index. if |
... |
arguments passed to
|
a topTable object produced by
topTable
, in addition to an
“SE” column containing the moderated standard
errors.
Mark Cowley, 2010-11-12
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.