The function PLNPCA() produces an instance of this class.
This class comes with a set of methods, some of them being useful for the user:
See the documentation for getBestModel(),
getModel() and plot().
The function PLNPCA(), the class PLNPCAfit()
PLNmodels::PLNfamily -> PLNPCAfamily
ranksthe dimensions of the successively fitted models
Inherited methods
new()Initialize all models in the collection.
PLNPCAfamily$new(
ranks,
responses,
covariates,
offsets,
weights,
formula,
control
)ranksthe dimensions of the successively fitted models
responsesthe matrix of responses common to every models
covariatesthe matrix of covariates common to every models
offsetsthe matrix of offsets common to every models
weightsthe vector of observation weights
formulamodel formula used for fitting, extracted from the formula in the upper-level call
controllist controlling the optimization and the model
optimize()Call to the C++ optimizer on all models of the collection
getModel()Extract model from collection and add "PCA" class for compatibility with factoextra::fviz()
varvalue of the parameter (rank for PLNPCA, sparsity for PLNnetwork) that identifies the model to be extracted from the collection. If no exact match is found, the model with closest parameter value is returned with a warning.
indexInteger index of the model to be returned. Only the first value is taken into account.
a PLNPCAfit object
getBestModel()Extract best model in the collection
PLNPCAfamily$getBestModel(crit = c("ICL", "BIC"))crita character for the criterion used to performed the selection. Either
"ICL", "BIC". Default is ICL
a PLNPCAfit object
plot()Lineplot of selected criteria for all models in the collection
PLNPCAfamily$plot(criteria = c("loglik", "BIC", "ICL"), reverse = FALSE)criteriaA valid model selection criteria for the collection of models. Any of "loglik", "BIC" or "ICL" (all).
reverseA logical indicating whether to plot the value of the criteria in the "natural" direction (loglik - penalty) or in the "reverse" direction (-2 loglik + penalty). Default to FALSE, i.e use the natural direction, on the same scale as the log-likelihood.
data(trichoptera)
trichoptera <- prepare_data(trichoptera$Abundance, trichoptera$Covariate)
myPCAs <- PLNPCA(Abundance ~ 1 + offset(log(Offset)), data = trichoptera, ranks = 1:5)
#>
#> Initialization...
#>
#> Adjusting 5 PLN models for PCA analysis.
#> Rank approximation = 5
Rank approximation = 2
Rank approximation = 3
Rank approximation = 1
Rank approximation = 4
#> Post-treatments
#> DONE!
class(myPCAs)
#> [1] "PLNPCAfamily" "PLNfamily" "R6"