Rnits (version 1.6.2)

calculateGCV: Calculate the optimal B-spline model using generalized cross-validation

Description

Calculate the optimal B-spline model using generalized cross-validation

Usage

calculateGCV(object, topcomp = 5)
"calculateGCV"(object, topcomp = 5)

Arguments

object

Rnits object

topcomp

The number of top eigenvectors to be used for computation

Value

A list object with fields 'degree', 'df' for each time series data set.

Details

The optimal B-spline model is chosen as the largest model that minimizes the cross validation error of the top N eigenvectors of each time series data.

Examples

# load pre-compiled expressionSet object for Ronen and Botstein yeast chemostat data
data(yeastchemostat)
rnitsobj = build.Rnits(yeastchemostat, logscale = TRUE, normmethod = 'Between')
opt_model <- calculateGCV(rnitsobj)
## Not run: 
# rnitsobj <- fit(rnitsobj, gene.level = TRUE, model = opt.model)
# ## End(Not run)