confint.hisemit: Extract Wald-type asymptotic confidence intervals from a hisemit object

Description

Extract Wald-type asymptotic confidence intervals from a hisemit object.

Usage

# S3 method for hisemit
confint(object, parm = c("lfdr", "fpp", "beta", "scale.fact", 
        "sd.ncp", "r", "coef", "pi0", "f"), level = 0.95, component, ...)

Arguments

object

A hisemit object.

parm

One of c("lfdr", "fpp", "beta", "scale.fact", "sd.ncp", "r", "coef", "pi0", "f"). See details.

level

A numeric scalar between 0 and 1, specifying the level of confidence.

component

Specifying which additive component to be extracted. See details.

…

Currently not used.

Value

A matrix (or vector) with columns giving lower and upper confidence limits for each parameter.

Details

The parm could be:

'lfdr' Local false discovery rates
'fpp' False positive proportions; not implemented yet.
'beta' Regression coefficients
'scale.fact' Multiplicative scale factor
'r' log(scale.fact-1)
'sd.ncp' Equivalent standard deviation of noncentrality parameters
'coef' All parameters
'pi0' Prior probability of true null hypotheses
'f' Underlying smooth function

The component specifies which component of the fitted smooth function to be extracted. If component is missing, the overal function is returned. If component='intercept', only the intercept term is returned. If component is numeric, then it specifies the smooth function corresponding to which covariate to be extracted.

References

Long Qu, Dan Nettleton, Jack Dekkers (2012) A hierarchical semiparametric model for incorporating inter-gene relationship information for analysis of genomic data. Biometrics, 68(4):1168-1177