MAINT.Data (version 0.5.1)

lda-methods: Methods for Function lda in Package `MAINT.Data'

Description

Linear Discriminant Analysis

Usage

"lda"(x,prior="proportions",selmodel=BestModel(H1res(x)),egvtol=1.0e-9,...)
"lda"(x, grouping, prior="proportions", tol=1.0e-4, 
subset=1:nrow(x) ,Config=1:5,SelCrit=c("AIC","BIC"))

Arguments

x

An IdtClMANOVA object representing the results of a (homocedastic normal-based) MANOVA of Interval Data, or an IData object representing NObs interval-valued entities

grouping

Factor indicating the group to which each observation belongs to

prior

The prior probabilities of class membership. If unspecified, the class proportions for the training set are used. If present, the probabilities should be specified in the order of the factor levels

selmodel

The index of model of IdtClMANOVA object to be used in the construction of the discriminant functions

tol

A tolerance to decide if a matrix is singular; it will reject variables and linear combinations of unit-variance variables whose standard deviation is less than tol

egvtol

A tolerance for positive eigenvalues of the scatter matris; eigenvalues below egvtol will be considered equal to 0, and the corresponding linear discriminant functions will be ignored

subset

An index vector specifying the cases to be used in the training sample. (NOTE: If given, this argument must be named

Config

Configuration of the variance-covariance matrix: a vector of integers between 1 and 5

SelCrit

The model selection criterion

...

Further arguments passed to or from other methods.

Value

Methods

signature(Idt = "IData")

Performs Linear Discriminant Analysis on interval-valued data

See Also

Idtlda, DACrossVal, qda

Examples

# Create an Interval-Data object containing the intervals for 899 observations 
# on the temperatures by quarter in 60 Chinese meteorological stations.
ChinaT <- IData(ChinaTemp[1:8])

#Linear Discriminant Analysis, assuming that Mid-Points are independent of LogRanges 
#    (Configuration C4)
ChinaldaC4 <- lda(ChinaT,ChinaTemp$GeoReg,Config=4)
cat("China linear discriminant analysis results =\n") ; print(ChinaldaC4)
cat("lda Prediction results:\n")
print(predict(ChinaldaC4,ChinaT)$class)

#Estimate error rates by three-fold cross-validation, replicated five times
CVldaC4 <- DACrossVal(ChinaT,ChinaTemp$GeoReg,TrainAlg=lda,Config=4,kfold=3,CVrep=5)
summary(CVldaC4[,,"Clerr"])
glberrors <- 
	apply(CVldaC4[,,"Nk"]*CVldaC4[,,"Clerr"],1,sum)/apply(CVldaC4[,,"Nk"],1,sum)
cat("Average global classification error =",mean(glberrors),"\n")