build_PCC_linear_models = function(x, y, seed_value=456){
    #
    # Builds (using caret's train function) several of the linear models discussed in
    # this chapter of the book optimizing the probability of correct classification PCC (i.e. accuracy).
    #
    # X [n samples, n features] needs to have zero variance columns removed (use nearZeroVar to remove them) and
    # linear dependent columns removed (use findLinearCombos to remove them)
    #
    # y [n samples, 1] is a two factor vector of predictions with the FIRST factor corresponding to the event of interest
    #
    # Written by:
    # -- 
    # John L. Weatherwax                2009-04-21
    # 
    # email: wax@alum.mit.edu
    # 
    # Please send comments and especially bug reports to the
    # above email address.
    #
    #-----

    # Linear Discriminant Analysis:
    #
    set.seed(seed_value)
    lda.classifier = train( X, y, method="lda", preProc=c("center","scale") )
    y_hat = predict( lda.classifier, X )
    cm = confusionMatrix( data=y_hat, reference=y )

    lda=list( classifier=lda.classifier, confusionMatrix=cm )

    # Penalized Methods:
    #
    glmnGrid = expand.grid(.alpha=c(0, 0.1, 0.2, 0.4, 0.6, 0.8, 1.0), .lambda=seq( 0.01, 0.2, length=40))
    set.seed(seed_value)
    glmnet.classifier = train( X, y, method="glmnet", tuneGrid=glmnGrid, preProc=c("center","scale") )
    y_hat = predict( glmnet.classifier, X )
    cm = confusionMatrix( data=y_hat, reference=y )
    glmnet=list( Classifier=glmnet.classifier, confusionMatrix=cm )

    # Nearest Shrunken Centroids:
    #
    nscGrid = expand.grid(.threshold=0:25)
    set.seed(seed_value)
    nsc.classifier = train( X, y, method="pam", tuneGrid=nscGrid, preProc=c("center","scale") )
    y_hat = predict( nsc.classifier, X )
    cm = confusionMatrix( data=y_hat, reference=y )
    nsc=list( classifier=nsc.classifier, confusionMatrix=cm )

    return( list( lda=lda, glmnet=glmnet, nsc=nsc ) )

}