nIC.est: Estimators based on model selection criteria.

In Shrinkage: Several Shrinkage Effect-Size Estimators

Description

Estimators based on a model selection criteria: Bayes factor, Akaike information or Bayesian information criteria.

Usage

nIC.est(x, y = NULL, opt = c('BF','AIC','BIC'), param0 = NULL, param = NULL,
  logx = TRUE, ...)

nBF_estimator(x, y = NULL, param0 = NULL, param = NULL, logx = TRUE, ...)

nAICc_estimator(x, y = NULL, param0 = NULL, param = NULL, logx = TRUE, ...)

nBIC_estimator(x, y = NULL, param0 = NULL, param = NULL, logx = TRUE, ...)

Arguments

x	Input data matrix: features(rows) x samples (columns). See examples.
y	Optional input data matrix.
opt	Option for selecting the type of estimator, it is a character: "BIC" estimator based on the Bayesian information criterion (BIC). Equivalent function: nBIC_estimator "AIC" estimator based on the Akaike information criterion corrected for small samples (AICc). Equivalent function: nAICc_estimator "BF" estimator based on the Bayes factor (BF). Equivalent function: nBF_estimator.
param	Numeric vector, the effect-size of the parameter of interest. If input param = NULL, it is internally computed from the input matrices x and y if they are given.
param0	Value of the effect-size of the parameter of interest corresponding to the null hypothesis (null value)(i.e. log fold change corresponding to no change, usually 0). If input param0 = NULL, it is internally set.
logx	If logx = TRUE (and param = NULL and param0 = NULL), param is computed internally considering that input matrices x and y are logarithms and thus param0 is set to 0.
...	Further arguments to pass to an internal function.

Value

A vector of length equal to the total number of features (i.e. proteins, genes,...).

Note

When inputs param and/or param0 are not given, they are computed internally from matrices x and y. If logx = TRUE then param = \bar{x} - \bar{y} and param0 is set to 0, while if logx = FALSE then param = \bar{x} / \bar{y} and param0 is set to 1.

Author(s)

Code: Zahra Montazeri, Corey M. Yanofsky, David R. Bickel and Marta Padilla (modifications)
Documentation: Alaa Ali and Marta Padilla

References

Yanofsky, C. M., & Bickel, D. R. (2010). Validation of differential gene expression algorithms: Application comparing fold-change estimation to hypothesis testing. BMC Bioinformatics, 11, 63.

Examples

#simulate some data sets: matrices of log-abundance levels
nsam<-5         #number of individuals
nfeat<-6        #number of features (metabolites, genes,...)
diffs<-c(1,4)   #features with differential log-abundance levels
lfc<-5          #differential quantity

# create xprnSet, xprnSetPair and numeric objects:
x <- matrix(runif(nfeat*nsam), nrow = nfeat, ncol = nsam) #case
y <- matrix(runif(nfeat*nsam), nrow = nfeat, ncol = nsam) #control
x[diffs,] <- x[diffs,] + lfc

# examples: ----------
z1 <- nIC.est(x=x,opt='BIC')
z2 <- nIC.est(x=x,opt='BF')   
z3 <- nIC.est(x=x,opt='AIC') 
z4 <- nIC.est(x=x,y=y,opt='BIC')
z5 <- nIC.est(x=x,y=y,opt='BF')   
z6 <- nIC.est(x=x,y=y,opt='AIC') 

Shrinkage documentation built on Sept. 12, 2016, 9:41 a.m.