cvq2.sample.B:
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 $x$.
Format
A data frame with six observations and one parameter per observation.
x- parameter
y- observation
Source
Generic data set, created for this purpose only.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 ($nRun=2,3\ldots, x$) 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.