criteria.values.G {MOODE} | R Documentation |
Evaluating individual criteria of the designs, from the Generalized compound criteria (Goos et al. 2005; Egorova,~O. 2017)
Description
Calculating values of determinant- and trace-based components of Generalized D-, DP-, L- and LP- criteria for an output of a search object, with model and control parameters set in a mood object.
Usage
criteria.values.G(search.obj, mood.obj, eps = 10^-23)
Arguments
search.obj |
Output of the ‘Search’ function |
mood.obj |
Output of the ‘mood’ function |
eps |
Computational tolerance, default 10^-20 |
Value
List of the calculated values:
-
df
pure error degrees of freedom -
Ds
Ds-criterion value, intercept excluded -
DP
DPs-criterion value, intercept excluded -
LoFD
LoF(D)-criterion value from the GD-criterion -
LoFDP
LoF(DP)-criterion value from the GDP-criterion -
biasD
bias(D)-criterion value from the GD-criterion -
Ls
Ls-criterion value, intercept excluded -
LP
LPs-criterion value, intercept excluded -
LoFL
LoF(L)-criterion value from the GL-criterion -
LoFLP
LoF(LP)-criterion value from the GLP-criterion -
biasL
bias(L)-criterion value from the GL-criterion
References
Egorova,~O. (2017).
Optimal Design of Experiments for Multiple Objectives.
Ph.D. thesis, University of Southampton.
Goos P, Kobilinsky A, O'Brien TE, Vandebroek M (2005).
“Model-Robust and Model-Sensitive Designs.”
Computational Statistics and Data Analysis, 49, 201-216.