cvq2.sample.B {cvq2}R Documentation

Small data set to demonstrate the difference between the conventional and the predictive squared correlation coefficient while performing a cross-validation.

Description

Contains a small data set with six observations, the observed value y depends on the parameter x.

Usage

data(cvq2.sample.B)

Format

A data frame with six observations and one parameter per observation.

x

parameter

y

observation

Details

The sample can be used to demonstrate the various settings of cvq2. The cross-validation applied to determine q^2 can be performed either as Leave-One-Out (nFold = N = 6) or as k-fold (nFold = 2,3).
In case nFold = 2,3 modelData is randomly split into nFold disjunct and equal sized test sets. Furthermore one has the opportunity to repeat the cross-validation, while each run (nRun=2,3…, x) has an individual test set compilation.
The prediction power, q^2_cv, calculated for this data set is considerably smaller than the model calibration, r^2, promises.

Source

Generic data set, created for this purpose only.

See Also

cvq2, q2


[Package cvq2 version 1.2.0 Index]