cvq2.sample.A {cvq2} | R Documentation |

##
Small data set to demonstrate the difference between the conventional and the predictive squared correlation coefficient.

### Description

Contains a small data set with four observations, the observation *y* depends on two parameters (*x_1*, *x_2*).

### Usage

data(cvq2.sample.A)

### Format

A data frame with four observations.
Each row contains two parameters and the observed value.

`x1`

parameter 1

`x2`

parameter 2

`y`

observation

### Details

This data set can be used to demonstrate the differences between the model calibration and the prediction power.
The prediction power can be determined either with cross-validation or the application of the model to the data set `cvq2.sample.A_pred`

.

### Note

This data set contains one outlier (row #2).
If the prediction power is determined with cross-validation, this outlier leads to a considerably decreased prediction power, *q^2_cv*, in comparison to the model calibration, *r^2*.
For this data set, one can perform a Leave-One-Out cross-validation only.

### Source

Generic data set, created for this purpose only.

### See Also

`cvq2`

,
`q2`

[Package

*cvq2* version 1.2.0

Index]