q2-class: Class '"q2"'

In cvq2: Calculate the predictive squared correlation coefficient

Description

The class "q2" is used to store information about the calibration of model M and its prediction performance. To determine the prediction power, M is applied to an external, independent data set.

Objects from the Class

Objects can be created by calls of the form new("q2", ...).

Slots

result

Contains two lists (fit, pred) regarding the results from linear regression (model calibration, fit) and the application of the model to a validation set (prediction power, pred)

output

A list of parameters like number formats, output restrictions or output targets

Model calibration

This part contains the measurements regarding the model calibration of the linear model M.

data

The observations and the linear fitted predictions by model M

data.col

The explanation of data's column names

model

The linear model M

n

The number of elements in the data set

observed_mean

The arithmetic mean of the observations

r2

The conventional squared correlation coefficient

rmse

The root mean square error with regard to the degree's of freedom ν

nu

The degree's of freedom

Prediction performance

This part contains the measurements regarding the prediction power of model M which ia applied to an external data set.

data

Contains the observations and their predictions by M

data.col

The explanation of data's column names

nTrainingSet

The number of elements in the model set (N-k)

nTestSet

The number of elements in the prediction set (k)

q2

The predictive squared correlation coefficient

rmse

The root mean square with regard to the degree's of freedom ν

nu

The degree's of freedom

Methods

show

Returns a comprehensive overview about the model calibration and the prediction performance.

Author(s)

Torsten Thalheim <[e-mail]>

Examples

  showClass("q2")

cvq2 documentation built on May 2, 2019, 8:29 a.m.